2275 Commits

Author SHA1 Message Date
Lourenço Vales
8e5b52cdc1 final fixes 2025-10-05 21:52:46 +02:00
Lourenço Vales
2182e2a4ea small typo 2025-10-05 21:36:38 +02:00
Lourenço Vales
ce4d2bfe50 changed validation for MX record corner case 2025-10-05 21:34:37 +02:00
Lourenço Vales
1787e44503 added more robust validation; small changes 2025-10-05 21:22:48 +02:00
Lourenço Vales
bc4f7b7309 fixed slight error in validation 2025-10-05 20:59:27 +02:00
Lourenço Vales
f7d8b42c7b adding doc comments 2025-10-05 20:49:02 +02:00
Lourenço Vales
1fb3ef1e71 fixed record deletion by changing the matching condition 2025-10-05 10:55:50 +02:00
Lourenço Vales
3e153f7f44 adding Poll condition to guarantee minimum time between tries 2025-10-05 10:39:47 +02:00
Lourenço Vales
8b30f7bd3d everything is implemented, now on to testing 2025-10-03 15:00:35 +02:00
Lourenço Vales
d304dafeea added partial cloudflare api integration 2025-10-03 09:29:32 +02:00
Lourenço Vales
9271906435 added cmp function 2025-10-03 09:29:32 +02:00
Lourenço Vales
7146139ae8 added CheckApply function; made some changes to structure 2025-10-03 09:29:32 +02:00
Lourenço Vales
ef74ede862 engine: resources: Add Cloudflare DNS resource 2025-10-03 09:29:32 +02:00
James Shubin
56e55dfad7 engine: resources: Sysctl was missing the errors 2025-10-03 01:34:26 -04:00
James Shubin
303e80dee7 engine, lang, lib, pgraph: Plumb through more ctx 2025-10-02 23:28:51 -04:00
James Shubin
6c6c9df75e engine: resources: Add a nif command and make it clearer
Let's have the opposite, and have an example so you don't forget which
is which.
2025-10-02 20:48:36 -04:00
James Shubin
5f4ae05340 readme: We moved to matrix 2025-10-02 03:05:05 -04:00
James Shubin
c48b884d16 misc: Add fpm repo script 2025-09-30 02:55:29 -04:00
James Shubin
fe77bce544 misc: Update old email address 2025-09-30 02:42:43 -04:00
James Shubin
26640df164 test: shell: Get the first ethernet device
In CI sometimes there are two, so this fails.
2025-09-30 00:27:35 -04:00
James Shubin
debd4ee653 misc: Remove old prototype language 2025-09-29 23:15:45 -04:00
James Shubin
63269fe343 spec: Small RPM fixes
In case anyone tries this.
2025-09-29 23:15:16 -04:00
James Shubin
f588703474 engine: resources: Add a gsettings resource
This adds a way to run the gsettings command for configuring dconf
settings usually used by GNOME applications.
2025-09-29 21:40:06 -04:00
James Shubin
52fbc31da7 engine: resources: Remove this noise 2025-09-29 21:26:14 -04:00
James Shubin
154f900d2a engine: resources: Add an ifequals option to block cmd
If the ifcmd returns true and this option is set, it will match that
output against this field, and if they match, then we skip cmd.

Much cleaner than needing to invoke bash to compare two strings.
2025-09-29 21:24:40 -04:00
James Shubin
bbd4f1dea1 engine: Add a utility function to copy the Init struct 2025-09-29 21:23:17 -04:00
James Shubin
22120649e5 modules: misc: Add simple flatpak management 2025-09-26 23:25:42 -04:00
James Shubin
a840dd43dd cli, tools, util, modules: Add a grow util and module
This builds in some functionality for growing the filesystem for new
machines. It also gets wrapped with an mcl module for ease of use.
2025-09-26 23:00:20 -04:00
James Shubin
83743df3e4 util: Add variant of exec cmd that returns output 2025-09-26 21:18:54 -04:00
James Shubin
15b2ff68cc lang: core: os: Simplify waitgroup
Doesn't need to be part of the struct. Maybe there are others like this
that need porting.
2025-09-26 21:18:54 -04:00
James Shubin
17544e881c test: Fix tag failures 2025-09-25 02:23:14 -04:00
James Shubin
6090517830 releases: Add release notes for 1.0.0 2025-09-25 02:11:05 -04:00
James Shubin
6a7b3d5fa9 readme: Ten 2025-09-25 01:16:03 -04:00
James Shubin
25804c71df lang: core embedded: provisioner: Work with USB-free machines
This feature was for machines that boot from USB keys. When we PXE boot,
this should not fail when the file isn't missing.
2025-09-16 01:11:32 -04:00
James Shubin
a54553c858 engine: resources: Print a warning if svc is slow
The biggest horror is blocked execution somewhere, so if the svc start,
stop or reload is being slow, then at least print a message to warn us.
2025-09-15 04:05:04 -04:00
James Shubin
ff1581be87 engine: resources: Massively refactor the svc
This was a long time coming, but now it looks to be done. It was kind of
meant as low-hanging fruit for some interested student, but in the end I
got to it first.
2025-09-15 04:05:04 -04:00
James Shubin
ec48a6944c lang: core: iter: Make map coding more consistent with filter
Keeping the code looking similar makes it easy to patch bugs that occur
in both.
2025-09-14 23:52:22 -04:00
James Shubin
df9849319d lang: core: iter: Replace graph when list length changes
The map and iter functions weren't replacing the graph if the input list
length changed. This was just an oversight in coding AFAICT, as the
sneaky case is that if the length stays the same, but the list contents
change, then it's okay to not swap.
2025-09-14 23:50:21 -04:00
James Shubin
045aa8820c engine: resources: Display tick marks for input range
This makes it prettier. We should also add the values, but this is
harder to do nicely.
2025-09-14 20:52:45 -04:00
James Shubin
a66cbc3098 engine: resources: Work around race in upstream lib
This is actually fixed in: 7d147928ee
but this is not in a release yet.
2025-09-14 19:36:57 -04:00
James Shubin
9833cb8df3 modules: virtualization: Update to Fedora 42 2025-09-14 00:06:59 -04:00
James Shubin
a73dc19ce9 engine: resources: Fix virt hotplug
At some point, this seems to have rotted, since I assume upstream
started requiring this updated XML spec. Fix it now.
2025-09-13 23:54:46 -04:00
James Shubin
bcf57f8581 engine: resources: Make the qemu guest agent automatic 2025-09-13 23:54:46 -04:00
James Shubin
611cdb3193 engine: resources: Disable buggy restart code
This was really not ever tested properly, and I worry it will deadlock.
It definitely kicks off falls positives that don't even do any harm as
far as we can tell.
2025-09-13 23:54:24 -04:00
James Shubin
1b39a780e1 engine: resources: Clean up virt code
There was and still is a bunch of terrible mess in this code. This does
some initial cleanup, and also fixes an important bug!

If you're provisioning a vmhost from scratch, then the function engine
might do some work to get the libvirt related services running before
the virt resource is used to build a vm. Since we had connection code in
Init() it would fail if it wasn't up already, meaning we'd have to write
fancy mcl code to avoid this, or we could do this refactor and keep
things more logical.
2025-09-13 23:28:33 -04:00
James Shubin
d59ae2e007 engine: graph: We shouldn't complain on context cancellation
These are expected from our engine. We do care about timeout's and so
on. This allows os to return ctx.Err() whenever a <-ctx.Done() happens,
which is more idiomatic for what we really want, but which we weren't
thorough with before.
2025-09-13 23:28:33 -04:00
James Shubin
b9363a3463 go: Update systemd dep to fix race
Hopefully this race is fixed upstream. Let's see.
2025-09-11 23:22:45 -04:00
James Shubin
a5f89d8d7b lang: funcs: dage: Use lazy freshness check
Not sure if this would introduce a glitch or not. Does seem to work
correctly. Without this, examples/lang/datetime2.mcl don't update
properly.
2025-09-11 23:19:45 -04:00
James Shubin
790b7199ca lang: New function engine
This mega patch primarily introduces a new function engine. The main
reasons for this new engine are:

1) Massively improved performance with lock-contended graphs.

Certain large function graphs could have very high lock-contention which
turned out to be much slower than I would have liked. This new algorithm
happens to be basically lock-free, so that's another helpful
improvement.

2) Glitch-free function graphs.

The function graphs could "glitch" (an FRP term) which could be
undesirable in theory. In practice this was never really an issue, and
I've not explicitly guaranteed that the new graphs are provably
glitch-free, but in practice things are a lot more consistent.

3) Simpler graph shape.

The new graphs don't require the private channels. This makes
understanding the graphs a lot easier.

4) Branched graphs only run half.

Previously we would run two pure side of an if statement, and while this
was mostly meant as an early experiment, it stayed in for far too long
and now was the right time to remove this. This also means our graphs
are much smaller and more efficient too.

Note that this changed the function API slightly. Everything has been
ported. It's possible that we introduce a new API in the future, but it
is unexpected to cause removal of the two current APIs.

In addition, we finally split out the "schedule" aspect from
world.schedule(). The "pick me" aspects now happen in a separate
resource, rather than as a yucky side-effect in the function. This also
lets us more precisely choose when we're scheduled, and we can observe
without being chosen too.

As usual many thanks to Sam for helping through some of the algorithmic
graph shape issues!
2025-09-11 23:19:45 -04:00
James Shubin
1e2db5b8c5 gapi: New API
Clear out lots of cruft and mistakes in this old API. The language GAPI
isn't updated with this commit, and as a result, this won't build, but
the fixes for it are coming shortly. We could have merged the two
commits, but it's easier to show them separately for clarity.
2025-09-09 02:21:59 -04:00
James Shubin
6041c5dc22 puppet: langpuppet: Nuke due to porting difficulties
Porting of the GAPI will cause challenging refactoring for me, so I'm
removing this for now, but happy to have it back if someone wants to
port it.
2025-09-09 02:21:59 -04:00
James Shubin
a668cd847e util: New buffered infinite chan primitive
I'm sure there are better implementations, but this feels clean enough
for now. Let's see if this is useful or not.
2025-09-09 02:21:59 -04:00
James Shubin
474df66ca0 lang: types: Add nil type for placeholder
The nil type is being added, but only slightly. It is not meant for real
use in the language. It will not work in all situations, and it isn't
implemented for many methods.

It is being added only for being a "dummy" placeholder value in the
engine when we have an unused value being propagated along an edge.
2025-09-09 02:21:59 -04:00
James Shubin
2022a31820 make: Leave race detector on by default
Maybe this will help us shake out some bugs.
2025-09-09 02:21:59 -04:00
James Shubin
71756df815 mod: Update fsnotify
This should fix a race condition in that library.

Likely fixed in: https://github.com/fsnotify/fsnotify/pull/678
2025-09-09 02:21:59 -04:00
James Shubin
f808c1ea0c converger: Wrap atomic lookup
Avoid this race. Maybe this code should be revisited with a mutex.
2025-09-09 02:21:59 -04:00
James Shubin
6c206b8010 util: Prevent unlikely race on easy exit
Race detector hit this up once, and I can see how it would be possible.
2025-09-09 02:21:59 -04:00
James Shubin
fb8958f192 engine: graph: Add err mutex
Here's a race that pops up. This is suboptimal locking, but it's not
important for now.
2025-09-09 02:21:59 -04:00
James Shubin
a070722937 etcd: Lock around read to prevent race 2025-09-09 02:21:59 -04:00
James Shubin
b02363ad0d etcd: scheduler: Use atomic to prevent race
This code should be rewritten, but in the meantime, at least avoid the
race detector issues.
2025-09-09 00:04:22 -04:00
James Shubin
bed7e6be79 etcd: Pass through the namespace
This is a bit tricky, and we should nuke and redo some of this API. The
sneaky bit has to do with whether we've already added the namespace
magic into our etcd client or not.
2025-09-09 00:04:22 -04:00
James Shubin
0031acbcbc lang: funcs: structs: Map indexes use half the integers
We want the pattern to be key:0, val:0, key:1, val:1, and so on... This
was previously using 0,1,2,3...

When we use Call directly, we need to fix this. Previously this was dead
code which is why the bug wasn't caught.
2025-09-09 00:04:22 -04:00
James Shubin
4e523231d6 engine: graph: Avoid race on fast pause variable
This code is basically unused, but let's keep it in for now in case we
eventually replace it with some contextual ctx code instead.
2025-09-09 00:04:22 -04:00
James Shubin
05d72b339d converger: Combine two signal channels into one
There's no reason we need to remake these two channels, when we can just
use one. We should probably rewrite this code entirely, but at least we
get rid of this race for now.
2025-09-09 00:04:21 -04:00
James Shubin
d2cda4ca78 etcd: Disable the dynamic chooser
We're not using dynamic etcd right now, so disable this code and prevent
the race detector complaining.
2025-09-09 00:04:21 -04:00
James Shubin
2f860be5fe engine: graph: Lock around frequent read races
These are all "safe" in terms of not ever conflicting, but the golang
memory model weirdness requires an actual lock to avoid race detector
errors.
2025-09-09 00:04:21 -04:00
James Shubin
5692837175 lang: Add a simple test of a non-tree dag 2025-09-09 00:04:21 -04:00
James Shubin
04ff2a8c5c lang: ast: Turn this speculation flag into a const
Makes it easier to find when debugging.
2025-09-09 00:04:21 -04:00
James Shubin
166b463fa9 lang: funcs: structs: Update the graph shape docs 2025-09-09 00:04:21 -04:00
James Shubin
2e858ff447 test: Improve colon test comment
I use these patterns when early hacking, and it's good to have a test to
catch them all before I merge.
2025-09-09 00:04:21 -04:00
James Shubin
6fac46da7c misc: Improved stack filtering
Although this needs more debugging, I'm not sure how the format changed.
I guess this is part of the "API" that golang is allowed to break ;)
2025-09-09 00:04:21 -04:00
James Shubin
2b820da311 lang: ast: structs, funcs: structs: Exprif without a channel
This adds an improved "expr if" which only adds the active branch to the
graph and removes the "secret" channel.
2025-08-04 17:45:06 -04:00
James Shubin
86c6ee8dee lang: ast, funcs: Remove the secret channel from call
This removes the secret channel from the call function. Having it made
it more complicated to write new function engines, and it's not clear
why it was even needed in the first place. It seems that even the
current generation of function engines work just fine without it.

Co-authored-by: Samuel Gélineau <gelisam@gmail.com>
2025-08-04 17:04:01 -04:00
James Shubin
0a76910902 lang: core: Skip broken test
I expect this will be deprecated soon, let's see.
2025-08-04 17:04:01 -04:00
James Shubin
138ff8a895 lang: funcs: structs: Fix typos 2025-07-17 01:47:55 -04:00
James Shubin
8edb8e2a7b lang: interfaces: Add new helpers for dealing with args 2025-07-17 01:47:55 -04:00
James Shubin
bdf5209f68 util: errwrap: Add unwrapping for context removal
It's common in many concurrent engines to have a situation where we
collect errors on shutdown. Errors can either because a context closed,
or because some engine error happened. The latter, can also cause the
former, leading to a list of returned errors. In these scenarios, we
want to filter out all the secondary context errors, unless that's all
that's there. This provides a helper function to do so.
2025-07-16 23:48:37 -04:00
James Shubin
299b49bb17 util: errwrap: Add a function for joining
This is like the Append function but for a list.
2025-07-16 23:48:37 -04:00
James Shubin
71e4282d3f lang: interfaces: Add args to struct helper
We should consider if it's possible to avoid all of this transforming
entirely, but at least for now, do it all in one place by having an
available helper.
2025-07-13 03:18:23 -04:00
James Shubin
984aa0f5fc lang: Rename the vertex names
Make it a bit more obvious what the generated nodes are for.
2025-07-13 03:18:23 -04:00
James Shubin
737d1c9004 lang: interfaces: Table can be a standalone type
We'd like to have some useful helpers defined on it, like Copy.
2025-07-13 03:18:23 -04:00
James Shubin
d113fcb6d7 lang: ast, interfaces, interpret: Table should be a well-known type
We use this in enough places, that it's nice to have it as a well-known
alias.
2025-07-13 03:18:23 -04:00
James Shubin
73e641120f pgraph: Improve time complexity of IncomingGraphVertices
This goes from O(n^2) to O(n) when map lookup is O(1). I never really
focused on much optimizing, but I noticed this one in passing.
2025-07-13 03:18:22 -04:00
James Shubin
f7e446ef6f lang: core: example: nested: We don't use the func suffix anymore
There are only functions, no need to add suffixes to file names.
2025-07-13 03:18:22 -04:00
James Shubin
21917864db lang: core, funcs: Remove facts API
This started because it was possible, not because it was very useful.
The overhead of using the full function API, is lessened by the function
API helpers, and the upcoming improvements in the function API.

It's much easier to have one fewer API's to manage and so on.

It's also a stark reminder of how weak tools like "puppet" are which
only really have data collection systems that don't take arguments.
2025-07-13 03:15:53 -04:00
James Shubin
c49d469dcd engine: resources: Work around trailing slash issue in home dir
If the user is logged in, and we try to change from /home/james to
/home/james/ we'll get the error:

usermod: user james is currently used by process ????

and furthermore, it makes no sense to try and make this change since the
usermod function won't do anything if you run:

usermod --gid james --groups wheel --home /home/james/ james

when /etc/passwd has /home/james as the string.
2025-06-25 06:05:28 -04:00
James Shubin
0a79daf277 pgraph: Print cycles on error
I'm a terrible algorithmist, so who knows if this is correct, but it
seems to work in my cursory testing.
2025-06-25 06:05:28 -04:00
James Shubin
a4ed647d02 modules: purpleidea: Add more packages 2025-06-25 06:05:28 -04:00
James Shubin
79c199975d modules: purpleidea: Add another useful helper package 2025-06-25 06:05:11 -04:00
James Shubin
50b4a2a4f7 lang: ast: Make error message clearer 2025-06-25 04:50:47 -04:00
James Shubin
f778008929 modules: misc: Key generation should support other types
I think these short keys are sketchy, but what do I know.
2025-06-25 04:50:47 -04:00
James Shubin
54380a2a1f modules: cups: Add more edges
Useful for performance reasons until we make autoedges blazing fast.
2025-06-25 04:50:47 -04:00
James Shubin
a5fc1256e2 lang: core: embedded: provisioner: Encrypt the filesystem
The provisioner should be able to encrypt things. We should use an empty
passphrase so that the choosing of the actual passphrase can be done at
first boot.
2025-06-23 19:53:52 -04:00
James Shubin
0b2236962c lang: core: embedded: provisioner: Separate home is rare
Maybe one day we want this off to prevent storage issues, but not today.
2025-06-12 18:17:33 -04:00
James Shubin
ee7ad7cbbe lang: core: embedded: provisioner: Skip ignore if no drives available
Small bug for certain setups.
2025-06-12 18:11:14 -04:00
James Shubin
7ba4c4960b modules: meta: Wrong interface for the loc network
Copy pasta bug!
2025-06-08 21:27:04 -04:00
James Shubin
777ea6115b lang: core: embedded: provisioner: Support exec handoff
Could be used for any tool, but mgmt is an obvious possibility.

I should check this code more, but it's roughly right and I'm sure it
will get refactored more when I build opt-in provisioning and so on.
2025-06-08 21:16:54 -04:00
James Shubin
582cea31b0 modules: Get rid of unnecessary printing 2025-06-08 04:33:35 -04:00
James Shubin
c107240098 modules: shorewall: Add manual edges for performance
If you don't want to use auto-edges, then this still works.
2025-06-08 04:30:39 -04:00
James Shubin
6265a330bf modules: misc: The COPR setup must be non-interactive 2025-06-08 04:21:58 -04:00
James Shubin
cfcb35456f modules: misc: Use a template for network
Just a bug, now fixed.
2025-06-08 04:21:27 -04:00
James Shubin
1ef7c370e7 etcd, engine: Fix typos 2025-06-08 03:36:11 -04:00
James Shubin
f22ec07ed3 lang: Improve logging of startup information
Graph size and build time are both helpful.
2025-06-08 03:36:11 -04:00
James Shubin
f594799a7f etcd: ssh: Improve the authentication for ssh etcd world
This was rather tricky, but I think I've learned a lot more about how
SSH actually works. We now only offer up to the server what we can
actually support, which lets us actually get back a host key we have a
chance of actually authenticating against.

Needed a new version of the ssh code and had to mess with go mod
garbage.
2025-06-08 03:07:59 -04:00
James Shubin
1ccec72a7c cli, etcd, lib, setup: Support ssh hostkey logic
This makes it easy to pass in the expected key so that we never have to
guess and risk MITM attacks.
2025-06-07 17:55:41 -04:00
James Shubin
55eeb50fb4 lang: Refactor all the highlight helping together
Keep this cleaner and add a bit more.
2025-06-07 17:52:15 -04:00
James Shubin
2b7e9c3200 engine, integration, setup: Seeds should be called properly 2025-06-07 17:52:15 -04:00
James Shubin
25263fe9ea cli: Allow multiples of these args
We forgot the flag that lets the CLI parser actually let us use
multiples.
2025-06-06 23:53:42 -04:00
James Shubin
1df28c1d00 lang: ast, funcs: Start plumbing through the textarea
We need to get these everywhere and this is a start.
2025-06-06 03:11:06 -04:00
James Shubin
32e91dc7de lang: interpolate: Add temporary textarea info to interpolation
We should really be doing the math to find out how far along the string
each token really is, but that's complicated and tedious, especially
with the simplification passes, so let's skip that for now and just show
the whole thing.
2025-06-06 03:11:06 -04:00
James Shubin
c2c6cb5b6a lang: interfaces: Subtle fixes in textarea
Turns out we need a lot more tests to make this work easier.
2025-06-06 03:11:06 -04:00
James Shubin
58461323b9 lang: parser: Try to add the end values in parser
Not sure if this is right, but it's a start.
2025-06-06 03:11:06 -04:00
James Shubin
cdc6743d83 lang: ast, interfaces, interpolate: Remove the legacy pos
This ports things to the new textarea. We need to plumb through things a
lot more, especially the string interpolation math to get the right
offsets everywhere, but that's coming.
2025-06-06 02:37:43 -04:00
James Shubin
86dfa5844a lang: ast: Add missing initialization calls
Not sure how we forgot these before.
2025-06-06 02:35:20 -04:00
James Shubin
5d44cd28db lang: ast, interpolate: Pass through uninterpolated strings
We don't need to make a new reference for nothing.
2025-06-06 01:00:11 -04:00
James Shubin
4f977dbe57 lang: Use the source finder wherever we can
This was easy to add and it works great!
2025-06-06 01:00:11 -04:00
James Shubin
573bd283cd lang: funcs: dage: Print out some error locations
Most things don't support this yet, but let's get in some initial
plumbing. It's always difficult to know which function failed, so we
need to start telling the users more precisely.
2025-06-06 01:00:11 -04:00
James Shubin
6ac72974eb lang: ast, interfaces: Move textarea to a common package
We're going to use it everywhere. We also make it more forgiving in the
meanwhile while we're porting things over.
2025-06-06 01:00:11 -04:00
James Shubin
4189a1299a lang: ast: Add scope feedback for classes
We did this elsewhere for functions, let's add classes too.
2025-06-06 01:00:11 -04:00
James Shubin
dcd4f0709f lang: ast: Provide better error reporting for scope errors
Print these file and line numbers when we can!
2025-06-05 23:00:56 -04:00
James Shubin
75bafa4fd3 mcl, docs: Use the less ambiguous form of the import
Update the style guide as well!
2025-06-05 22:47:38 -04:00
James Shubin
e5ec13f592 modules: misc: Add a class to install a copr repo 2025-06-05 22:47:38 -04:00
James Shubin
1a0fcfb829 modules: misc: Use the less ambiguous import name 2025-06-05 22:34:33 -04:00
James Shubin
ba86665cbb misc: Rename for consistency 2025-06-05 22:34:29 -04:00
James Shubin
301ce03061 misc, setup, util: Add a ulimit
I think this gives us the ability to open more files.
2025-06-05 22:34:29 -04:00
James Shubin
650e8392c5 golang: Tidy mod file
I ran go mod tidy.
2025-06-05 21:46:28 -04:00
James Shubin
d7534b2b3b golang: Update go.mod to avoid tidy warnings 2025-06-05 21:45:06 -04:00
James Shubin
3b88ad3794 lang: core: os: Add a modinfo function
Not sure if this will need renaming, but might be a useful family of
functions.
2025-06-05 21:40:19 -04:00
James Shubin
499b8f2732 lang: funcs: dage: Make error clearer
The implementation of the specific function is sending a nil value,
which is not allowed. This is a bug in the code of that function.
2025-06-05 20:36:15 -04:00
James Shubin
ac3a131a9f modules: meta: Improve firewall rules for our router 2025-06-05 14:47:46 -04:00
James Shubin
a72492f042 modules: meta: Move router to networkd
I can't get NetworkManager working properly in parallel to wireguard. I
get an extra route added and it breaks the tunnel. No idea why. The
networkd equivalent seems to just work.
2025-06-05 14:47:46 -04:00
James Shubin
c51a55e98a modules: misc: Add networkd helpers 2025-06-05 14:47:46 -04:00
James Shubin
892fd1e691 modules: misc: Add a network manager dhcp interface 2025-06-05 14:47:46 -04:00
James Shubin
23aa18d363 modules: shorewall: Refactor to allow bulk rules
Very useful for brownfield deployments where we're migrating a ton of
rules over.
2025-06-05 14:47:46 -04:00
James Shubin
d14930ef28 lang: core: embedded: provisioner: Don't provision to USB disks
This should hopefully skip over any USB drives. Of course if we actually
want to provision to a USB drive, then we'll have to add a feature flag
for that.
2025-06-05 14:47:46 -04:00
James Shubin
81063ae6df etcd: ssh: Reconnect on SSH failures
If the SSH connection dies, the dialer can now reconnect that portion.
2025-06-05 14:47:46 -04:00
James Shubin
f42daf4509 etcd: ssh: Improve logging to be less misleading 2025-06-05 14:47:46 -04:00
James Shubin
1caf6fb3bf etcd: ssh: Pass through the ctx into the SSH dialer
I hope I did this correctly.
2025-06-05 14:47:46 -04:00
James Shubin
16ade43caf engine: Rename world API and add a context
We want to be able to pass ctx through for various reasons.
2025-06-05 14:47:46 -04:00
James Shubin
99d8846934 engine: resources: Remove env
The nix people want this sometimes, and don't in others ;)

Karpfen, you now owe me a $beverage =D
2025-06-05 14:47:09 -04:00
Lourenço Vales
2d78dc9836 modules: contrib: Add cryptpad module
This is a module to deploy cryptpad locally - as it is, this is only
intended for dev purposes, not for production environments
2025-05-26 14:28:33 +02:00
James Shubin
b85751e07e setup: Add the ssh url arg to the setup utility 2025-05-26 02:26:32 -04:00
James Shubin
0fd6970c0a engine: resources: The http server flag res should autogroup
If we want to receive more than on flag (key) value, then these
obviously need to autogroup together, because it's the same http server
request that comes in and which should be shared by everyone with the
same path.
2025-05-25 04:46:34 -04:00
James Shubin
936cf7dd9d engine: graph: autogroup: Ensure we sort correctly
This should ensure this sort order of longer thing first, when running
the algorithm. We want to autogroup two or more http:server:flag
resources together first, before the whole grouped resource gets pulled
into the http:server one.
2025-05-25 04:21:44 -04:00
James Shubin
fd5bc63293 engine: traits: Allow self-kind with hierarchical grouping
Not sure why I was blocking this previously, perhaps there was no use
case, and I was trying to ensure there wasn't accidental, unwanted
grouping? Perhaps I wanted to improve performance? In any case, let's
turn this off for now, and check for bugs.
2025-05-25 04:19:38 -04:00
James Shubin
be4cb6658e engine: graph: Simplify the Send/Recv code
This code might be slightly redundant, so simplify it. If something went
wrong, this will be the commit that did it, but it seems relatively
safe.
2025-05-25 04:18:26 -04:00
James Shubin
efff84bbd4 engine: graph: autogroup: Print these errors when debugging
This gives important clues as to why something isn't grouping as
expected. Show them if needed.
2025-05-25 03:48:57 -04:00
James Shubin
74f36c5d73 engine: resources: Add some compile time checks for groupers
These can "break" silently and not autogroup if we change the resource
and it no longer fulfills the interface. Add this compile time check to
prevent that.
2025-05-25 03:47:47 -04:00
James Shubin
b868a60f69 engine: resources: Simplify the Watch loop
I had some legacy unnecessary boolean for sending everywhere. Not sure
why I never re-read it, it's so easy to just copy and paste and carry
on.
2025-05-25 02:12:14 -04:00
James Shubin
f73127ec23 engine: resources: Make error not ambiguous
The same text exists elsewhere.
2025-05-25 01:42:41 -04:00
James Shubin
654e958d3f engine: resources: Add the proper prefix to grouped http resources
Resources that can be grouped into the http:server resource must have
that prefix. Grouping is basically hierarchical, and without that common
prefix, it means we'd have to special-case our grouping algorithm.
2025-05-25 01:40:25 -04:00
James Shubin
1f54253f95 engine: resources: Add a trim field to line resource 2025-05-25 01:40:21 -04:00
James Shubin
2948644536 lang: ast: Remove the double dash
Not sure why this is here?
2025-05-25 01:17:28 -04:00
James Shubin
d2403d2f0c etcd: client: str: We do not want the prefix match
This was a likely copy+pasta error, since we match precise strings here.
If we had two similarly prefixed strings, we'd have an error.
2025-05-25 01:17:28 -04:00
James Shubin
876834ff29 lang: core: fmt: Catch printf edge case 2025-05-25 01:17:28 -04:00
James Shubin
861ba50f9c engine: resources: Add a ui redirect
I always forget the /index.html part so make it easier!
2025-05-15 02:52:57 -04:00
James Shubin
43492a8cfa make: Add missing clean target for wasm 2025-05-15 02:41:58 -04:00
James Shubin
287504cfa8 engine: resources: Add missing struct tags to kv 2025-05-15 01:46:03 -04:00
James Shubin
0847b27f6a modules: misc: Add a pattern for systemd daemon reload
It would be really nice if systemd actually had an API for getting
events on this.
2025-05-14 21:21:27 -04:00
James Shubin
aa4320dd5f modules: misc: Add some authorized key work
More testing and features are needed, but this is a good start.
2025-05-09 04:11:04 -04:00
James Shubin
7c5adb1fec lang: core: net: Add a helper to return the network ip 2025-05-09 02:49:02 -04:00
James Shubin
20e1c461b8 lang: core: embedded: provisioner: Update import style 2025-05-09 02:48:51 -04:00
James Shubin
e9d485b7f6 lang: ast, core: Add some safety checks
I don't think I'm hitting these, but good for debugging.
2025-05-09 01:08:12 -04:00
James Shubin
e86d66b906 engine: resources: Avoid double slash on error
Errors will include a second slash if this ends with one. Might as well
clean it to avoid the semblance of a bug.
2025-05-09 00:00:46 -04:00
James Shubin
9a63fadfbd engine: resources: Rename var
So it doesn't conflict with "path" import.
2025-05-08 23:16:57 -04:00
James Shubin
7afa372765 engine: resources: Let the user race me
If a user is racing the file resource, don't error permanently, just
skip the file that vanished, and move on with your life.
2025-05-08 23:11:51 -04:00
James Shubin
fddebb2474 engine, lang: core: Match exported resources properly
I inverted the logic for complex setups and forgot to handle the zero
cases. I also didn't notice my loop continue error. This cleans all this
up so that we can have proper exported resource matching.
2025-05-08 22:29:03 -04:00
James Shubin
ad0dd44130 engine: Don't force validation for hidden resources
I think this is what I want in most scenarios, is there a reason to do
otherwise? This is because we may wish to export incomplete resources,
where the remaining necessary fields for validation happens on collect.
2025-05-06 03:36:01 -04:00
James Shubin
2ee403bab9 git: Update gitignore files
We were overly matching in some cases by not starting with a slash. This
updates a few other cases too.
2025-05-06 02:52:26 -04:00
James Shubin
0e34f13cce engine: resources: Add a line resource
Simple enough for the common cases. It just needs some tests.
2025-05-06 02:22:39 -04:00
Lourenço Vales
f2a6a6769f engine: resources: Add a WatchFiles field to exec
This adds a field that takes a list of files for exec to watch for
events on.
2025-05-05 23:54:33 -04:00
James Shubin
4903995052 lang: core: os: Add readfilewait which won't error easily
Just a useful helper function which we may want for a while.
2025-05-05 23:54:28 -04:00
James Shubin
774d408e13 engine: Fix up some send/recv corner cases
Initially I wasn't 100% clear or decided on the send/recv semantics.
After some experimenting, I think this is much closer to what we want.
Nothing should break or regress here, this only enables more
possibilities.
2025-05-05 23:53:37 -04:00
James Shubin
ae1d9b94d4 engine: util: Add a debug utility
This is useful for some patches. Let's see if I can remember to use and
improve it!
2025-05-05 22:30:31 -04:00
James Shubin
267bcc144b engine: util: Clean up error messages 2025-05-05 22:29:10 -04:00
James Shubin
fd40c3b64f engine: util: Fix grammar typo 2025-05-05 20:21:29 -04:00
James Shubin
e2b6da01d8 engine: graph: Fix messy imports 2025-05-05 20:21:29 -04:00
James Shubin
dad15f6adc examples: lang: Add missing folder 2025-05-04 13:54:45 -04:00
James Shubin
6ec707aea7 examples: lang: Simplify a common example 2025-05-02 03:12:18 -04:00
James Shubin
807c4b3430 engine: resources: Add an http ui resource
Many years ago I built and demoed a prototype of a simple web ui with a
slider, and as you moved it left and right, it started up or shutdown
some number of virtual machines.

The webui was standalone code, but the rough idea of having events from
a high-level overview flow into mgmt, was what I wanted to test out. At
this stage, I didn't even have the language built yet. This prototype
helped convince me of the way a web ui would fit into everything.

Years later, I build an autogrouping prototype which looks quite similar
to what we have today. I recently picked it back up to polish it a bit
more. It's certainly not perfect, and might even be buggy, but it's
useful enough that it's worth sharing.

If I had more cycles, I'd probably consider removing the "store" mode,
and replace it with the normal "value" system, but we would need the
resource "mutate" API if we wanted this. This would allow us to directly
change the "value" field, without triggering a graph swap, which would
be a lot less clunky than the "store" situation.

Of course I'd love to see a GTK version of this concept, but I figured
it would be more practical to have a web ui over HTTP.

One notable missing feature, is that if the "web ui" changes (rather
than just a value changing) we need to offer to the user to reload it.
It currently doesn't get an event for that, and so don't confuse your
users. We also need to be better at validating "untrusted" input here.

There's also no major reason to use the "gin" framework, we should
probably redo this with the standard library alone, but it was easier
for me to push out something quick this way. We can optimize that later.

Lastly, this is all quite ugly since I'm not a very good web dev, so if
you want to make this polished, please do! The wasm code is also quite
terrible due to limitations in the compiler, and maybe one day when that
works better and doesn't constantly deadlock, we can improve it.
2025-05-02 02:14:14 -04:00
James Shubin
6b10477ebc lang: core: convert: Add a simple str to int function 2025-05-02 00:24:21 -04:00
James Shubin
412e480b44 engine: local: Get the logic right
I think we were not benefitting from the cache and sending unnecessary
events. It would be great to have tests for this, but commit this fix
for now, and be embarrassed in the future if I got this code wrong.
2025-05-02 00:04:00 -04:00
James Shubin
cc2a235fbb engine: resources: Add a reminder about events
I might want to do this some day, it could be important. Look into it.
2025-05-02 00:04:00 -04:00
James Shubin
7c77efec1d engine: resources: Cleanup this old code
This is equivalent and cleaner.
2025-05-02 00:04:00 -04:00
James Shubin
4b1548488d lib: It is called mcl officially for a while now 2025-04-28 00:31:14 -04:00
James Shubin
47aecd25c3 lang: funcs: structs: Pass through the type
Not sure why this wasn't done or if it should be, but seems plausible
for now.
2025-04-27 22:23:42 -04:00
James Shubin
fb6eae184a lang: ast: Refactor to unindent slightly 2025-04-27 22:19:14 -04:00
James Shubin
16d3e3063c lang: funcs: facts: Do not reuse fact pointers
In my carelessness, I was re-using pointers when a fact was used twice!
This could cause disastrous consequences like a double close panic on a
datetime.now() fact for example.

In other news, we should consider if it's possible to get more clever
about graph shape optimization so that we don't need more than once
instance of certain functions like datetime.now() in our graph.
2025-04-27 22:14:51 -04:00
James Shubin
37bb67dffd lang: Improve graph shape with speculative execution
Most of the time, we don't need to have a dynamic call sub graph, since
the actual function call could be represented statically as it
originally was before lambda functions were implemented. Simplifying the
graph shape has important performance benefits in terms of both keep the
graph smaller (memory, etc) and in avoiding the need to run transactions
at runtime (speed) to reshape the graph.

Co-authored-by: Samuel Gélineau <gelisam@gmail.com>
2025-04-27 22:14:51 -04:00
James Shubin
9c9f2f558a lang: Move out this legacy execution function
Hasn't been used in a while, but it's fine if we want to use it for
tests.
2025-04-22 03:24:23 -04:00
James Shubin
1a81e57410 lang: interfaces: Update stale comments 2025-04-22 03:24:23 -04:00
James Shubin
7096293885 lang: funcs: dage: Return better errors
Helps a lot with debugging.
2025-04-22 03:24:23 -04:00
James Shubin
1536a94026 lang: Functions that build should be copyable
It's not entirely clear if this is required, but it's probably a good
idea. We should consider making it a requirement of the BuildableFunc
interface.
2025-04-22 03:24:23 -04:00
James Shubin
1bb1e056c4 lang: funcs: structs: Add some extra safety checks
Not sure if these are even needed.
2025-04-22 03:24:23 -04:00
James Shubin
e71b11f843 lang: funcs: facts: Check if a fact is callable 2025-04-22 03:24:23 -04:00
James Shubin
b4769eefd9 lang: funcs: facts: Add a separate callable interface
Add some symmetry to our interfaces for now, even though I'd love to
drop the idea of "facts" altogether.
2025-04-22 03:24:23 -04:00
James Shubin
d4a24d4c9d lang: funcs: wrapped: Simplify the implementation 2025-04-22 03:24:23 -04:00
Ahmad Abuziad
c5d7fdb0a3 util: Add a bunch of tests
This improves our test coverage significantly.
2025-04-22 03:18:49 -04:00
Lourenço Vales
ae68dd79cb lang: core: iter: Add a range function
This commit implements a range function that mimicks python's range
built-in by having a start, stop, and range argument. There's also
a few examples and tests to mimick Python's examples to guarantee
we're consistent with their behaviour.
2025-04-22 02:37:35 -04:00
James Shubin
de970ee557 engine: resources: Add symlink param to the file res
This adds initial symlink support to the file resource, and while it is
hopefully correct, there are always sneaky edge cases around symlinks
and security, so review and tests are highly encouraged!
2025-04-22 02:21:58 -04:00
James Shubin
60a3d7c65e lang: interfaces: Add more information about graph semantics
Sam thoughts.
2025-04-19 13:02:51 -04:00
James Shubin
9c1c587f7b lang: parse, core: world: Add a collect package
Some checks failed
/ Test (basic) on ubuntu-latest with golang 1.23 (push) Has been cancelled
/ Test (race) on ubuntu-latest with golang 1.23 (push) Has been cancelled
/ Test (shell) on ubuntu-latest with golang 1.23 (push) Has been cancelled
This lets us look at the available resource data for collection, and to
filter it so we can decide what we want to collect on our machine.

Other types of collect functions could be added in the future.
2025-04-05 17:00:53 -04:00
James Shubin
af04d364d0 lang: core: fmt: Make printf handle more cases
Until we make a clean determination about what this should print, this
should handle things for now.
2025-04-05 16:14:11 -04:00
James Shubin
748f05732a engine, etcd: Watch on star pattern for all hostnames
We forgot to watch on star hostname matches.
2025-04-05 15:45:44 -04:00
James Shubin
148bd50e9f engine, etcd: Prevent engine thrashing
These two small bugs would allow thrashing to occur since we'd
constantly delete and re-add exports, and constantly think that a noop
etcd operation made a change.
2025-04-05 15:28:54 -04:00
James Shubin
6c1c08ceda engine: resources: Test to make sure metaparams are preserved
We should ensure these get preserved across encoding/decoding. We rely
on this behaviour.
2025-04-05 12:45:23 -04:00
James Shubin
045b29291e engine, lang: Modern exported resources
I've been waiting to write this patch for a long time. I firmly believe
that the idea of "exported resources" was truly a brilliant one, but
which was never even properly understood by its original inventors! This
patch set aims to show how it should have been done.

The main differences are:

* Real-time modelling, since "once per run" makes no sense.
* Filter with code/functions not language syntax.
* Directed exporting to limit the intended recipients.

The next step is to add more "World" reading and filtering functions to
make it easy and expressive to make your selection of resources to
collect!
2025-04-05 12:45:23 -04:00
James Shubin
955112f64f engine: Let others use the ResUID struct
It's a useful key in maps.
2025-04-05 12:45:23 -04:00
James Shubin
7f341cee84 engine: resources: Improving logging even more
Messages should happen after the event on success. The error scenario
has its own pathway to report.
2025-04-05 12:45:23 -04:00
James Shubin
f71e623931 engine: resources: Print a message on empty file creation
We don't see this event happening which is confusing. There might be
other cases we didn't handle cleanly.
2025-04-05 12:45:23 -04:00
James Shubin
8ff187b4e9 lang: core, funcs: Rename things for consistency
Seems we had different patterns going on. This makes those all
consistent now.
2025-04-05 12:45:23 -04:00
James Shubin
30aca74089 engine, yamlgraph: Disable the old exported resources stuff
These were really just stubs so that I could prove out the reactive
model very early, and I don't think they're really used anywhere.

I'm also not really using the yamlgraph frontend. If someone wants any
of that code, step up, or it will rot even more.
2025-04-05 12:45:23 -04:00
James Shubin
3dfca97f86 engine: Add a method to determine if a res kind is valid 2025-04-05 12:45:23 -04:00
James Shubin
0d4c6e272d lib: Add timing for topological sort
At least for consistency with everyone else...
2025-04-05 12:45:23 -04:00
James Shubin
fce250b8af cli, etcd, lib: Fixup golint issues with SSH
This stuff is arbitrary and stupid.
2025-04-05 12:45:23 -04:00
James Shubin
f6a8404f9f modules: virtualization: Qcow2 should be the default
Snapshots and so much more is only possible with qcow2. A long time ago
it had performance issues, but things seem okay now.
2025-03-28 04:44:56 -04:00
Karpfen
c50a578426 git: Add vendor/ to gitignore 2025-03-22 14:56:16 -04:00
Karpfen
7e8ced534f misc: Use /usr/bin/env for a more generic shebang
Use path based SHELL in Makefiles. It was suggested that this is a
better solution for make for cases when there is no /usr/bin/env.

See: https://github.com/purpleidea/mgmt/pull/694#discussion_r1015596204
2025-03-22 14:53:21 -04:00
Lourenço Vales
f2d9219218 lang: core: os: Add is_virtual function
This is a basic implementation of a detection method for whether mgmt is
running in a virtualized environment. We achieve this by doing two types
of checks: on one hand, we check if the CPU flags can confirm the
presence of a virtualized env; on the other, we check if the presence
of known files related with DMI (and their contents) can confirm whether
 we're inside a virt env. Either of these situations will cause the
function to return true, with the default case being false. All of these
checks are relatively naive and can be improved by looking at the main
inspiration for this implementation, which was systemd's own check for
the presence of virtualization.
2025-03-21 14:18:55 -04:00
James Shubin
f269096eb9 cli, etcd, lib: Remove the etcd client from main
We are slowly getting rid of more cruft and abstracting it nicely. More
to go!
2025-03-19 06:01:42 -04:00
James Shubin
5665259784 cli, engine, etcd, lib: Move the hostname value to the API
Every world implementation needs a unique UUID, might as well move this
to the API.
2025-03-19 05:41:04 -04:00
James Shubin
02fca6409a cli, etcd, lib: Add an etcd client over ssh world backend
This provides a new kind of "world" backend, one that runs etcd over an
SSH connection. This is useful for situations where you want to run an
etcd cluster somewhere for clients across the net, but where you don't
want to expose the ports publicly.

If SSH authentication is setup correctly (using public keys) this will
tunnel over SSH for etcd to connect.

This patch does not yet support deploys over SSH, but that should be
fixed in the future as the world code gets cleaned up more.
2025-03-19 05:33:07 -04:00
James Shubin
a7a5237b07 cli, engine, etcd, lib: Pass in init args
Improve the API and make it more general.
2025-03-18 04:54:13 -04:00
James Shubin
7ad54fe3e8 cli, engine, etcd, lib: Split out the deployer into world
This should hopefully make the refactor into a clean world API a bit
better. Still more to do though!
2025-03-18 04:54:13 -04:00
James Shubin
1a35ab61ca engine: Split out the world filesystem interface 2025-03-18 03:32:42 -04:00
James Shubin
59c33a354c engine, lang: core: world: Split out the scheduler interface 2025-03-18 03:32:42 -04:00
James Shubin
c853e24ded engine: Split out the str world interface
This is a core API and it should really be cleaned up if possible.
2025-03-18 03:32:42 -04:00
James Shubin
692db084e4 engine: Split off resource world interface 2025-03-18 03:32:42 -04:00
James Shubin
1edff3b3f5 engine: Move another interface method 2025-03-18 03:32:42 -04:00
James Shubin
b173d9f8ef engine: Split out the etcd cluster size options
This is clean up work so that it's easier to generalize the world
backends.
2025-03-18 03:32:42 -04:00
James Shubin
a697add8d0 modules: dhcp: Add more package versions 2025-03-17 22:01:43 -04:00
James Shubin
c83e2cb877 git: Add a small ignore entry 2025-03-17 21:51:45 -04:00
James Shubin
642c6b952f lang: core, funcs: Port some functions to CallableFunc API
Some modern features of our function engine and language might require
this new API, so port what we can and figure out the rest later.
2025-03-16 23:23:57 -04:00
James Shubin
f313380480 engine: resources: Container stopped should be valid for no container 2025-03-13 01:03:11 -04:00
James Shubin
f8a4751290 engine: resources: Don't prematurely error docker watches
A subtlety about the engine is that while it guarantees CheckApply
happens in the listed edge-based dependency order, it doesn't stop
Watch from starting up in whatever order it wants to. As a result, we
can prematurely error since the docker service isn't running yet. It may
in fact be in the process of getting installed and started by mgmt
before we then try and use this resource! As a result, let it error once
for free and wait for CheckApply to get going before we start again.

Keep in mind, Watch has to use the .Running() method once to tell
CheckApply to do its initial event. So this concurrency is complex!

It's unclear if this is a bug in mgmt or not, but I'm leaning towards
not, particularly since there isn't an obvious way to fix it.
2025-03-12 06:14:38 -04:00
James Shubin
3ca1aa9cb1 engine: resources: Fix backwards docker ports
This wasn't setup properly, now it's fixed. Woops.
2025-03-12 05:45:27 -04:00
James Shubin
37308b950b cli, gapi: Add more information that deploy is running
There can be a non-obvious pause, so give some hint here...
2025-03-12 05:45:26 -04:00
James Shubin
05306e3729 engine: resources: Modernize the docker resources
They made the assumption that there would be a based docker service
installed at Init which could not be guaranteed. Also use the internal
metaparameter timeout feature instead of private counters.
2025-03-12 05:45:26 -04:00
James Shubin
a6057319a9 lang: Make scope error messages be more consistent 2025-03-12 03:33:08 -04:00
James Shubin
87d8533bd0 lib: Patch out the mess when using our magic option 2025-03-11 04:53:08 -04:00
James Shubin
dce83efa96 etcd: Add a special magic option hack
Workaround some legacy code for now.
2025-03-11 04:53:08 -04:00
James Shubin
1cb9648b08 etcd: Workaround possible rare deadlock
This code is terrible, but maybe this is good enough for now.
2025-03-11 04:18:03 -04:00
James Shubin
17b859d0d7 cli, gapi, lang, lib: Add a flag to skip autoedges
The GAPI API is a bit of a mess, but I think this seems to work for
standalone run and also deploys. Hopefully I didn't add any unnecessary
extra dead code here, but that's archaeology for another day.
2025-03-11 04:18:03 -04:00
James Shubin
8d34910b9b modules: prometheus: Fix title of service template 2025-03-11 04:18:02 -04:00
James Shubin
5667fec410 modules: prometheus: Remove erroneous tmpl extension 2025-03-11 04:18:02 -04:00
James Shubin
46035fee83 engine: resources: Add simple configuration steps to virt builder
This adds some simplistic configuration management / provisioning
functionality to this virt:builder resource which makes it easier to
kick off special functionality that we might want to build.
2025-03-11 04:18:02 -04:00
James Shubin
219d25b330 engine: resources, modules: virtualization: Add a seeds option
This makes it easier to configure the machine by giving it an automatic
initial setup of an mgmt client.
2025-03-11 04:18:02 -04:00
James Shubin
181aab9c81 engine: resources: Fix small cmp typo in virt builder res 2025-03-10 19:01:05 -04:00
James Shubin
aabcaa7c8c setup: Error if no options are specified 2025-03-10 18:31:20 -04:00
James Shubin
09f3b8c05f setup: Add seeds and no server feature
We will want both of these for most clustered setups.
2025-03-10 16:24:18 -04:00
James Shubin
f5e2fde20d cli, lib: Fix typo 2025-03-10 16:23:56 -04:00
James Shubin
50bd6f5811 lib, gapi, cli: Add a wait flag to empty and a new default
Change the default "wait" state for if you run the empty frontend when
there's already an available deploy waiting. You almost certainly want
to start running it right away.

Example:

mgmt etcd

mgmt run --hostname h1 --no-server --tmp-prefix --seeds=http://127.0.0.1:2379 empty
mgmt run --hostname h2 --no-server --tmp-prefix --seeds=http://127.0.0.1:2379 empty

mgmt deploy --no-git --seeds=http://127.0.0.1:2379 lang examples/lang/hello0.mcl

mgmt run --hostname h3 --no-server --tmp-prefix --seeds=http://127.0.0.1:2379 empty

In fact, you don't even need to start up etcd first for this to all
work.
2025-03-10 14:56:42 -04:00
James Shubin
37e5a37045 setup: Fix firstboot typo 2025-03-09 15:55:58 -04:00
James Shubin
8544a66257 lang: Allow more than one possible error in tests
There are some rare situations with completely symmetrical graphs which
mean that there isn't a "more correct" error. This is due to the
annoying map iteration non-determinism, and so instead of fighting to
remove every bit of that, let's just accept more than one error here.
2025-03-09 03:03:37 -04:00
James Shubin
a50765393d lang: ast: Catch ordering errors 2025-03-09 01:50:28 -05:00
James Shubin
6bae5fc561 pgraph: Make our slow toposort even slower
I think this makes it more deterministic, but I'm not sure it matters,
since we are comparing based in the .String() property, and some nodes
have the same value, so it ends up depending on the order they're added
to the graph datastructure, but then we lose this information since it's
a map. Yuck.
2025-03-09 01:50:28 -05:00
James Shubin
f87c550be1 lang: ast, interfaces: Improve speculation safety checks
We want to speculate in more cases, so make sure that speculation is
safe!
2025-03-08 17:45:29 -05:00
James Shubin
aea894a706 lang: ast: Add more context to pointer errors
This makes debugging easier. We don't expect these errors to occur with
normal usage.
2025-03-08 17:45:29 -05:00
James Shubin
a549a30f71 lang: ast: Add more context to table errors
This makes debugging easier. We don't expect these errors to occur with
normal usage.
2025-03-08 17:45:29 -05:00
James Shubin
2899bc234a lang: Add a forkv loop statement for iterating over a map
This adds a forkv statement which is used to iterate over a map with a
body of statements. This is an important data transformation tool which
should be used sparingly, but is important to have.

An import statement inside of a forkv loop is not currently supported.
We have a simple hack to detect the obvious cases, but more deeply
nested scenarios probably won't be caught, and you'll get an obscure
error message if you try to do this.

This was incredibly challenging to get right, and it's all thanks to Sam
for his brilliance.

Note, I couldn't think of a better keyword that "forkv" but suggestions
are welcome if you think you have a better idea. Other ideas were formap
and foreach, but neither got me very excited.
2025-03-08 17:45:29 -05:00
James Shubin
cf7e73bbf6 lang: Add a for loop statement for iterating over a list
This adds a for statement which is used to iterate over a list with a
body of statements. This is an important data transformation tool which
should be used sparingly, but is important to have.

An import statement inside of a for loop is not currently supported. We
have a simple hack to detect the obvious cases, but more deeply nested
scenarios probably won't be caught, and you'll get an obscure error
message if you try to do this.

This was incredibly challenging to get right, and it's all thanks to Sam
for his brilliance.

Co-authored-by: Samuel Gélineau <gelisam@gmail.com>
2025-03-08 17:45:29 -05:00
James Shubin
c456a5ab97 lang: types: Add some length methods for list and map 2025-03-06 16:55:55 -05:00
James Shubin
b5ae96e0d4 lang: types: Add some helpful true and false values
In case we need one, we don't need to build it.
2025-03-06 16:55:55 -05:00
James Shubin
f792facde9 lang: gapi: Debug stalled graph errors
This makes debugging these scenarios a lot easier.
2025-03-05 17:24:26 -05:00
James Shubin
a64e3ee179 misc: Add baddev tooling
I think this is the *wrong* way to build this, but it's perfectly legal
to have a feature branch with this committed that people can develop
against. We can always cherry-pick off those commits to merge them, and
we can update and rebase this commit over time when needed.
2025-02-27 17:18:39 -05:00
James Shubin
c5257dd64b lang: parser: Simplify code and format it
This would get done by gofmt -s anyways.
2025-02-27 17:13:31 -05:00
James Shubin
f74bc969ca make: Cleanup old targets 2025-02-27 17:01:37 -05:00
Edward Toroshchyn
63d7b8e51e engine: resources: exec: Fix wrong err variable being checked in test 2025-02-27 14:50:38 -05:00
James Shubin
d56896cb0d lang: core: Simplify list and map lookup default functions 2025-02-26 19:59:47 -05:00
James Shubin
d579787bcd lang: core: Simplify list and map lookup functions 2025-02-26 19:59:47 -05:00
James Shubin
37fffce9f5 lang: core: Simplify implementation of the "contains" function 2025-02-26 18:12:38 -05:00
James Shubin
d7ecc72b41 lang: ast, gapi, interfaces, parser: Print line numbers on error
This adds an initial implementation of printing line numbers on type
unification errors. It also attempts to print a visual position
indicator for most scenarios.

This patch was started by Felix Frank and finished by James Shubin.

Co-authored-by: Felix Frank <Felix.Frank.de@gmail.com>
2025-02-25 20:15:02 -05:00
James Shubin
f754bbbf90 git: Add more entries to gitignore file 2025-02-25 12:10:04 -05:00
Felix Frank
bb171ced86 misc: Add missing sudo invocation 2025-02-24 10:43:43 -05:00
Edward Toroshchyn
c25a2a257b misc: Fix typos and spelling errors 2025-02-24 16:01:46 +01:00
Lourenço
1f90de31e7 lang: core: net: Add a new func for URL parsing
This is a first attempt to add a new function for URL parsing, using
go's net/url package and the simple API. This is still a barebones
implementation, there's possibility to expose more information. It also
includes simple tests.
2025-02-19 13:35:20 +01:00
Lourenço
b5384d1278 engine: resources: Adding logic for svc unit state
This is a small patch that adds logic for checking what's the state of
the unit file and making the CheckApply function more robust
2025-02-18 17:14:27 -05:00
James Shubin
d80ec4aaa7 engine: resources: Detect simple self-referential frags
It would be a likely mistake to create a self-referential frag, and mgmt
would spin forever updating the file... We probably don't want this, so
let's just catch this case in Validate.

Of course you could get around this with multiple files, and a fancier
search could statically check the graph, but the goal isn't to prevent
any bad code, since that's not likely to be possible.
2025-02-15 06:58:15 -05:00
James Shubin
5d63376087 modules: meta: Remove duplicate line 2025-02-13 06:33:47 -05:00
James Shubin
4fd6ced287 docs: Add new talks from Belgium, 2025 2025-02-10 10:34:53 -05:00
James Shubin
82489c3fe0 engine: resources: Add shell field to user resource 2025-02-07 18:08:25 -05:00
James Shubin
a064a87ecd lang: Add a weird test case
Mark Smith was concerned we might not handle this case correctly. It
seems we do in fact catch this scenario, so it's not an issue. Yay!
2025-02-07 17:57:36 -05:00
James Shubin
f51a1200d1 util: Add a helper to get the users shell entry 2025-02-07 17:57:36 -05:00
James Shubin
ecd5a0f304 util, lang, etcd: Move the error type to our util package
We use this error in a lot of places, let's centralize it a bit.
2025-02-07 17:57:36 -05:00
James Shubin
096ef4cc66 engine: resources: Modernize the user resource
Do some small fixups like adding ctx and fixing obvious bugs.
2025-02-07 17:57:36 -05:00
James Shubin
7da98ef349 test: Rename the reflowed comments test to make it easier to find
This makes running one-of executions of this a bit easier.
2025-02-06 08:19:22 -05:00
hades
8cd7fa27e2 engine: resources: exec: Add a bit of documentation to exec res 2025-02-06 08:18:48 -05:00
James Shubin
134e2f1cd9 examples: lang: Add a new env example 2025-02-06 07:49:35 -05:00
Edward Toroshchyn
042ae02428 engine: resources: exec: Add tests to check env values 2025-02-06 07:14:12 -05:00
Edward Toroshchyn
1e33c1fdae misc: Add vim syntax highlighting file
This is an extremely basic initial version of syntax highlighting, written just
so that I can edit the MCL files in vim and not cry.

The following features are supported:
 - MCL keywords
 - strings (including escape characters)
 - comments
 - built-in resources (as of 0.0.27)
2025-02-05 08:50:50 -05:00
James Shubin
bdc46648ff modules: Add prometheus and grafana modules
These are really stubs, and need some more testing and integration, but
there were some people who expressed interest in this, so let's push it
early.
2025-02-03 04:46:44 -05:00
James Shubin
ab9c1d3d96 modules: cups: Fixup obvious missing bits
I didn't merge these parts because I have some other WIP code I was
working on. Might as well put this in now.
2025-02-02 01:51:41 -05:00
James Shubin
0fb546ad61 engine: resources: Make some svc cleanups
We would often actually drop the refresh because of bad checks.
2025-02-02 01:43:14 -05:00
James Shubin
7439d532c7 modules: dhcp: Include stub hosts file
Even if we don't have any hosts, we should still have a valid config.
2025-01-31 11:44:30 -05:00
James Shubin
de9c0adcc0 releases: Add release notes for 0.0.27
I did this build with: `make release` followed by:
GOTAGS='noaugeas' make release when the arm64 build failed.
2025-01-31 03:22:45 -05:00
James Shubin
e40819d617 modules: Add a small stub for CUPS
This is definitely not perfect, but it's a simple stub which we can
expand on.
2025-01-31 00:52:24 -05:00
James Shubin
7331d3a7ee util: pprof: Improve pprof for ease of use
This makes it slightly cleaner.
2025-01-31 00:00:05 -05:00
James Shubin
95f353c6a4 modules: virtualization: Add some simpler helpers
You don't need much to build a vm host. Here's the start.
2025-01-30 23:04:12 -05:00
James Shubin
5044ef4e8a engine: resources: Add a virt builder password selector
This is very helpful for debugging, particularly with broken Debian
installs.
2025-01-30 22:56:24 -05:00
James Shubin
3c61d088ab test: Make sure to test the docs
I can't believe I forgot to enable this earlier.
2025-01-29 12:18:55 -05:00
James Shubin
315a493565 lang: Add a few more tests 2025-01-26 19:48:17 -05:00
James Shubin
6268b61a7d lang: core: Lookup function (with default) can be more precise 2025-01-26 19:12:13 -05:00
James Shubin
3f202c6a7a lang: core: Fix struct lookup corner case
We forgot to reject this corner case which could lead to a runtime error
since the expected type from the incoming struct would not match what
we're handling.
2025-01-26 19:12:13 -05:00
James Shubin
d46c43df5a lang: core: Let lookup function specialize earlier
If we happen to know some information, we can specialize early and help
type unification solve things.
2025-01-26 18:21:54 -05:00
James Shubin
1538befc93 lang: ast, parser: Allow calling anonymous functions
I forgot to plumb this in through the parser. Pretty easy to add,
hopefully I didn't forget any weird corner scope cases here.
2025-01-26 17:21:11 -05:00
James Shubin
1af334f2ce misc: Update to golang 1.23
The 1.22 version has some issues that makes it more difficult to
workaround, so we'll move right away to 1.23
2025-01-26 17:17:02 -05:00
James Shubin
d30ea571f1 misc: Update to golang 1.22
Ran:

go get -u ./...
go mod tidy

We also got rid of travis and simplified things a bit.
2025-01-26 17:16:40 -05:00
James Shubin
d30ff6cfae legal: Remove year
Instead of constantly making these updates, let's just remove the year
since things are stored in git anyways, and this is not an actual modern
legal risk anymore.
2025-01-26 16:24:51 -05:00
James Shubin
1d3f2dbe3c util: Add a pprof helper package
As it turns out, there are other helper packages out there. I wrote this
before I realized that, and it didn't take me that long anyways.
2025-01-24 03:06:31 -05:00
James Shubin
ca6e7ad432 lang: types, lang: core: strings: Add a new join_nonempty function
Along with a helper function for when we need a []string from a Value.
2025-01-24 00:07:23 -05:00
James Shubin
f92afe9ae4 modules: meta: Add a router meta module
Useful for setting up simple routers. Some auto-provisioning polish
would definitely help, but this is pretty useful already.
2025-01-18 01:22:33 -05:00
James Shubin
483cc22c32 lang: core: embedded: provisioner: Add IPXE support
This lets you boot from ipxe. You can run the ipxe shell from their
stock image or the netboot.xyz one. For the latter, press "m", then type
"dhcp" (machine is now pingable!) then type "route" to check the ip.

To boot type:

chain http://192.168.42.1:4280/menu.ipxe

and you're off!

Thanks to frebib for finding the workaround to the VFS bug. The answer
is you need to run the imgfree command to unblock the initrd.
2025-01-18 01:07:19 -05:00
James Shubin
2f3bd72491 lang: core: embedded: provisioner: Fix regression in Fedora 41
Unfortunately with Fedora 41 and DNF5, this breaks what our host machine
expects. So you now need dnf5+ to make this work, or you can revert this
commit.
2025-01-18 01:02:17 -05:00
James Shubin
6499fcb1e0 modules: dhcp: Support the authoritative setting
Quite useful if we plumb it all the way through!
2025-01-18 00:30:46 -05:00
James Shubin
12a0600d38 engine: resources: Improve hostname change message
Make it clearer where empty strings are, and that it already happened.
2025-01-17 23:58:26 -05:00
James Shubin
cace2bacb8 lang: core: embedded: provisioner: Add support for module path
This is needed for bigger code bases. Remember: this is consulted at the
deploy stage, and the deploy contains the entire tree (including
modules) of everything it needs to run. This is why you don't need to
add a --module-path arg when running mgmt in systemd with the "empty"
frontend.
2025-01-17 23:54:30 -05:00
James Shubin
05d440114a lang: core: embedded: provisioner: Remove stale comment 2025-01-17 23:49:21 -05:00
James Shubin
b392285e1d gapi, lang: gapi: Make dir to avoid errors
It seems we sometimes need to make the intermediate dir.
2025-01-17 23:32:13 -05:00
James Shubin
a713c08585 engine: resources: Make net and firewalld resources quieter
They are too noisy!
2025-01-17 23:32:13 -05:00
James Shubin
8e8e831e73 modules: misc: Switch type to list of strings
The DNS entry should be a list of strings. We would have caught this
earlier but this helped us find a type unification issue.
2025-01-17 23:28:05 -05:00
James Shubin
86b95b2c0b modules: shorewall: Fix type on stoppedrules file 2025-01-17 22:12:55 -05:00
James Shubin
4a578ca40c lang: core: embedded: provisioner: Handoff a hostname
This makes it easier to run config tools if the hostname is already set.
2025-01-17 19:48:23 -05:00
James Shubin
a60148f370 lang: core: net: Add more ip utilities 2025-01-17 18:58:21 -05:00
James Shubin
00366de67b modules: purpleidea: Update my personal packages 2025-01-17 18:57:18 -05:00
James Shubin
a08ba0b0e9 engine: resources: Tar now accepts dirs without a trailing slash
If these are found, then the dir path itself is copied in as well.
2025-01-17 18:15:45 -05:00
James Shubin
81b102ed7f lang: ast: Allow multiple star imports
If more than one star import is present in the same scope, allow it. If
one star import could overwrite something, ordering is not guaranteed.
We allow this for now, but we might create a compiler fix to stop it.
This adds a test to notice both of these behaviours.
2025-01-17 14:03:48 -05:00
James Shubin
c8f911ec5d lang: core: embedded: provisioner: Fix panic
The conditions changed for this one, make it stricter.
2025-01-17 12:35:30 -05:00
James Shubin
7694da4241 lang: core: embedded: provisioner: Skip bmc if empty 2025-01-17 12:34:27 -05:00
James Shubin
a0d500a602 lang: interfaces: Remove dangerous init method
This can cause the source to get overwritten and changed and is usually
unnecessary.
2025-01-04 20:56:11 -05:00
James Shubin
553172992f lang: parser: Typo fix 2025-01-04 20:55:41 -05:00
James Shubin
e6d614f4dd engine: resources: Add a bmc resource
This resource manages bmc devices in servers or elsewhere. This also
integrates with the provisioner code.
2025-01-03 18:19:31 -05:00
James Shubin
3107dfbd08 modules: shorewall: Small fixups to improve the module
These are some common fixes and improvements for normal shorewall usage.
As we shake out more uses of this, we find small issues. This lets us
have long rules, and a better default config.
2025-01-02 15:43:27 -05:00
James Shubin
802823dcb0 modules: misc: Support VIP's in network config 2025-01-02 15:42:40 -05:00
James Shubin
5858c8b501 pgraph: Panic if vertex is nil
These should be caught early.
2024-12-18 13:49:13 -05:00
James Shubin
2561dba8f5 lang: Increase debugging timeout 2024-12-15 10:39:46 -05:00
James Shubin
f5806e0617 lang: funcs: structs: Add Call and CallStruct methods for composite
It might turn out the CallStruct is the API we want. This will depend on
the future iterations of the function engine.
2024-12-08 17:43:41 -05:00
James Shubin
e9dbb7b86c lang: funcs: structs: Add Call method for const 2024-12-08 16:48:04 -05:00
James Shubin
28f5b8331a lang: funcs: structs: Improve naming
These could print nicer for debugging.
2024-12-08 16:24:42 -05:00
James Shubin
5ff4f0456a lang: Add mode to overwrite tests
This is useful if we need to reformat a bunch of previously passing
tests.
2024-12-08 15:34:28 -05:00
James Shubin
82c614f2d9 engine: resources: Workaround broken debian package when building images 2024-12-06 15:59:39 -05:00
James Shubin
50265d2303 engine: resources: Pull out a distro function for guest 2024-12-06 15:58:51 -05:00
James Shubin
ecee84aa28 docs: New blog post about modules and imports 2024-12-03 03:00:21 -05:00
James Shubin
2e146e8c8e engine: resources: Fix up some issues with cron
This really needs looking at again, since it's so old and buggy or
broken.
2024-12-03 01:52:23 -05:00
James Shubin
097efdd66a engine: graph: autoedge: Clean up redundant logs
They repeat themselves, this is cleaner.
2024-12-03 00:56:22 -05:00
James Shubin
5764c977f1 modules: misc: Don't ignore the router setting
This mistake caused us to ignore the router setting when we wanted it!
Woops =D
2024-12-03 00:43:32 -05:00
James Shubin
4d30772b3b lang: Add test for unused include statements
Even if they have no side effect, they aren't legal, since it would be
surprising if suddenly something new got added in one which then broke
the imports.
2024-11-28 19:17:33 -05:00
James Shubin
8472b1ebf2 lang: gapi: Convenience the user by allowing relative dirs in cli 2024-11-28 19:17:17 -05:00
James Shubin
e1070d3e13 lang: ast, download: Improve error messages 2024-11-28 19:17:17 -05:00
James Shubin
98d7f294eb lang: download: Improve git reliability
Fix a small panic that could happen if we had a bad clone, and make
visibility into this operation better. We also make room for future
context cancellation since the library now supports this.
2024-11-28 16:38:33 -05:00
James Shubin
517fc1e05b lang: gapi: Remap the module path correctly
If we've set the --module-path arg, then we expect to rebase that path
out of the deploy, and instead it should show up as /modules/ which is
the standard. Handle this scenario.
2024-11-28 16:29:33 -05:00
James Shubin
c2f75d64a6 util: arch: The Any value should be the same everywhere
In 80e8c9cadc when this was ported, the
"Any" value diverged accidentally. This would cause some packages to not
be found, since they didn't match any arch.

Thanks to karpfen to digging into the issue.
2024-11-28 14:55:30 -05:00
James Shubin
380004b1dc readme, docs: New docs available 2024-11-23 01:25:40 -05:00
James Shubin
28a443d11d docs: Add a hack for golang functions 2024-11-22 14:20:24 -05:00
James Shubin
a600e11100 cli, docs: Add a docs command for doc generation
This took a lot longer than it looks to get right. It's not perfect, but
it now reliably generates documentation which we can put into gohugo.
2024-11-22 14:20:16 -05:00
James Shubin
7b45f94bb0 lang: core: Remove the unnecessary func suffix
We don't really need these, it's clear what things are.
2024-11-22 01:18:19 -05:00
James Shubin
acdd6476f2 test: Remove empty variable
Copy-pasta bug!
2024-11-21 23:49:32 -05:00
James Shubin
018d3efc90 lang: funcs: Move standalone functions into core
Everything should be all together.
2024-11-21 22:56:17 -05:00
James Shubin
b40d10a366 util: Add a generic map key and value swapping function
Fun little utility function which is useful.
2024-11-21 02:54:10 -05:00
James Shubin
a88034ab06 modules: misc: Add standard header 2024-11-20 23:47:17 -05:00
James Shubin
907d2ad1a1 modules: dhcp: Add an mgmt module for managing dhcpd
This is not perfect, but it's a good start, and it shows how a module
might be structured.
2024-11-18 15:12:01 -05:00
James Shubin
3bd6986fde modules: shorewall: Add an mgmt module for managing shorewall
This is not perfect, but it's a good start, and it shows how a module
might be structured.
2024-11-18 15:11:31 -05:00
James Shubin
43bd847bad modules: misc: Improvements on ip address setting 2024-11-08 14:12:02 -05:00
James Shubin
0c0583adc8 modules: misc: Add network manipulation helpers
This is common functionality which we might want to use on new machines.
2024-11-06 22:13:31 -05:00
James Shubin
c642b5eeae lang: core: net: Add new function to get cidr prefix 2024-11-06 21:10:04 -05:00
James Shubin
69e84fbbed engine: resources: cron: Add ctx where possible.
Lots of the API's here now support this. Here's an example, work on the
others too.
2024-11-06 21:09:50 -05:00
James Shubin
f8b06f32ec engine: resources: Remove unused wait group 2024-11-06 21:09:50 -05:00
James Shubin
59a20f53eb lang: core: sys, engine: resource: Update hostname functionality
We didn't have a solid resource and sys.hostname() didn't have events!
2024-11-06 21:09:50 -05:00
James Shubin
83fd8b7e54 engine: util: Add more cmp utility functions 2024-11-06 20:02:10 -05:00
James Shubin
098ab20ec9 lang: gapi: Duplicates are possible if we have a diamond dag
Allow this, just remove them...
2024-11-06 20:02:10 -05:00
James Shubin
a2ce9e890d lang: core: net: Add a way to get the machine mac addresses 2024-11-05 14:55:58 -05:00
James Shubin
be7a5399e3 lang: core: util: Add hostname mapper function
This adds a new util package with some useful functionality which could
be implemented as pure mcl, but instead we add it here as a good place
to help with code reuse.
2024-11-05 14:55:58 -05:00
James Shubin
3fb492f6aa util: Add a TLS helper
Make it easier to build TLS stuff in pure golang.
2024-11-01 19:41:35 -04:00
James Shubin
e4f062b006 engine: resources: Parse distro properly
I regressed here when patching this. Here's the fix. Would be beautiful
to have hardware to run end-to-end testing on. If you want to sponsor
this, please let me know!
2024-10-30 16:04:01 -04:00
James Shubin
422719c345 lang: core: map: Add functions to extract keys and values
Simple stuff, but now it's done!
2024-10-30 00:58:10 -04:00
James Shubin
71a1efde99 examples: tls: Add a simple TLS example
Was useful for testing things...
2024-10-30 00:41:26 -04:00
James Shubin
ed84c5460c lang: core: embedded: provisioner: Workaround bad mirrors
With the release of Fedora 41, I was getting lots of mirror errors.
Hopefully this helps make it more robust. It was failing repeatedly
while trying to download packages, and I kept having to restart things,
but once I added this option things worked. Hopefully they're related.
2024-10-29 18:53:58 -04:00
James Shubin
0222a682fc lang: core: embedded: provisioner: Not sure we need this package
Seems to cause some issues. Remove it for now.
2024-10-29 18:53:36 -04:00
James Shubin
1cd4af5838 lang: core: embedded: provisioner: Keep this message separate
Don't autogroup it with others.
2024-10-29 16:58:06 -04:00
James Shubin
d1aaf6e82b lang: core: embedded: provisioner: Handle spurious failures
Not sure why this happen, I think it's just random network blips. Simple
retry should be used for now.
2024-10-29 16:58:06 -04:00
James Shubin
52a71f9515 lang: core: embedded: provisioner: Add unused repo generation
I was playing around with generating repo's, but they didn't turn out to
be needed at this time. Committing this anyways for future reference.
2024-10-29 16:58:06 -04:00
James Shubin
3c665174cc lang: core: embedded: provisioner: Implement handoff
Here's a good first way to implement handoff. What's particularly
elegant about handoff here, is that this is the first form of it I know,
where handoff happens between a provisioning tool and a configuration
management tool and those are the same tool! As a result, this can allow
for some really elegant integration, and the end-user never has to deal
with the combinatorial explosion of the N * M scenario of gluing each
provisioning tool to each different configuration management tool.

We'll have other forms of handoff in the future, but this simple
approach is useful already.
2024-10-29 16:58:06 -04:00
James Shubin
93eb8b2b76 lang: core: embedded: provisioner: Host an available deploy
This makes the current deploy available. This is likely not useful when
this is used from the embedded provisioner cli tool, since it would
contain cli functions that won't run, but it is useful when it's used as
a library.
2024-10-29 16:42:15 -04:00
James Shubin
1692235498 lang: core: embedded: provisioner: Log output of post
This is useful for debugging and for knowing what we really did to the
machine.
2024-10-29 16:42:15 -04:00
James Shubin
a6bcd4b92b lang: core: embedded: provisioner: Reduce amount of log noise
We don't need to refresh the leases so often.
2024-10-29 16:42:15 -04:00
James Shubin
d065cddf5e lang: core: embedded: provisioner: Remove old package
Doesn't seem to be used for anything at the moment.
2024-10-29 16:42:15 -04:00
James Shubin
20d4809e8e engine: resources: Print netmask nicely for our DHCP resource
Makes it easier to see what's going on.
2024-10-29 16:42:15 -04:00
James Shubin
b074386c26 cli: Add setup and firstboot commands
This adds two new top-level commands: setup and firstboot.

Firstboot is pure-golang implementation of a service that runs some
commands once when a system first boots. You need to install this
service, and put the scripts to run in a special directory. This is
inspired by the virt-builder --firstboot mechanism.

Setup is a general purpose command that makes it easy to setup certain
facilities on a new machine. These include the mgmt package dependencies
it might need, a service to run it from, and the necessary service to
use the mgmt firstboot service as well.

All of this has been built to facilitate handoff between provisioning a
new machine and running configuration management on it.
2024-10-29 16:42:15 -04:00
James Shubin
b140b2dfeb util: Move executable path finding into a helper function 2024-10-29 16:41:37 -04:00
James Shubin
8e3d959500 util: We prefer to append rather than truncate
This makes this utility function more useful.
2024-10-29 16:41:37 -04:00
James Shubin
8c886bbe7c util: Nil input to our simple cmd helper should be allowed
Don't panic here!
2024-10-29 16:41:37 -04:00
James Shubin
7d204dfb74 util: Add a simple append file write function
Similar to the golang os lib version except we append.
2024-10-29 16:39:46 -04:00
James Shubin
583f90dc7b util: distro: Rename functions to avoid golang stutter warning 2024-10-25 02:56:12 -04:00
James Shubin
85e1d6c0e8 engine: resources: Make sure to set the netmask
Some clients would DECLINE if this was not set. This was reproduced my
using the stock coredhcp DHCPv4 server and disabling the netmask plugin.
One of the clients that would DECLINE is a Lenovo ThinkCentre m90n doing
a UEFI (PXE) netboot.

This was found in 1327752725 and is
hopefully now completely fixed!
2024-10-25 00:57:50 -04:00
James Shubin
2c967e3897 util: Add a simple template system for systemd unit files
Just the basics for what we need, nothing more. Not intended as a
general-purpose library for use elsewhere.
2024-10-24 17:29:00 -04:00
James Shubin
202a8e1fba util: Add a small helper to exec commands simply 2024-10-18 10:02:29 -04:00
James Shubin
e6085d77ff util: Add an flock utility for lock file type things
Useful for ensuring only one binary runs at the same time.
2024-10-18 10:02:29 -04:00
James Shubin
10f82c6566 lang: core: list: Add a concat function
It works with arbitrary numbers of arguments too!
2024-10-18 10:02:29 -04:00
James Shubin
3d11b2caaf lang: core: deploy: Add a function to help in obtaining bootstrap deps
This should make it easier to implement handoff.
2024-10-15 20:54:33 -04:00
James Shubin
f8037a1f99 lang: types: Add a small helper function for common type conversions 2024-10-15 20:53:39 -04:00
James Shubin
067eef9007 util: distro, engine: resources: Update virt-builder res
A little easier to maintain if we support more distros eventually.
2024-10-15 20:36:51 -04:00
James Shubin
e45d9be065 util: distro: Parse the os-release file 2024-10-15 20:36:51 -04:00
James Shubin
d24149518c util: distro: Refactor family and distro code
I hate writing abstraction code like this, but I'm hoping it will be
useful.
2024-10-15 20:36:50 -04:00
James Shubin
d403f18b2a util: distro: Put distro specific data in this util package
Try and see if we can put all our distro specific stuff in here...
2024-10-15 19:04:17 -04:00
James Shubin
1f12150d8f engine: resources, examples: lang: Expand on future deploy ideas
This is one idea for reference. But I doubt this is needed anytime soon
since we have a good working solution in the examples.
2024-10-13 20:47:47 -04:00
James Shubin
d3a7cefcc6 engine: resources: Add an archiving, deploy resource
This makes a bundle out of the code in the current deploy. Hopefully
this is useful for handoff!
2024-10-13 16:44:23 -04:00
James Shubin
a8c8f09aa3 gapi: Plumb through a URI mechanism
This is at least a stop-gap until we redo the whole filesystem API mess.
I think golang is partly to blame because they don't have proper API's
merged yet.
2024-10-13 16:40:50 -04:00
James Shubin
b03fdeccae lang: interfaces: Nil input means no args 2024-10-07 00:00:19 -04:00
James Shubin
6c12e8a29b lang: gapi: The module path needs separate rebasing
If we're using a module path which doesn't share the same root, rebase
it separately.
2024-10-07 00:00:19 -04:00
James Shubin
310452542b lang: gapi: Rebase to the common path prefix
When the modules dir is not within the main base, we don't correctly
choose the common base path. As a result, we should choose something
common for our internal path representation.
2024-10-05 01:02:03 -04:00
James Shubin
b514022713 util: Add some path manipulation algorithms
These could use some optimization by an algorithmist! Not urgent right
now since they're not currently in any fast paths in the code.
2024-10-05 01:02:03 -04:00
James Shubin
c937280664 engine: local, lang: core: local: Add a pool function
This adds a new local API for pool allocation, and with it a
corresponding function in the "local" import.
2024-10-04 22:37:39 -04:00
James Shubin
898b58e3e7 lang: core: strings: Add a substring function named substr 2024-10-03 13:49:08 -04:00
James Shubin
74119a0a53 lang: core: strings, util: Add left and right padding functions
Golang didn't want these in the standard library, but they are useful.
2024-10-03 13:49:08 -04:00
James Shubin
d6914d3437 lang: core: net: Improve formatting of the mac functions 2024-10-03 13:49:08 -04:00
James Shubin
fdfa03685c lang: core: net: Add is_mac function 2024-10-02 14:31:20 -04:00
James Shubin
149a85fcde modules: misc: Ensure the ssh key has a folder 2024-09-29 00:53:11 -04:00
James Shubin
65f26769ae lang: core: net: Add a cidr to mask function
And tidy things up slightly to remove buggy tests.
2024-09-28 22:54:06 -04:00
James Shubin
6397c8f930 lang: Clean up import logs
We get a lot of useless noise here, make it neater.
2024-09-28 21:56:53 -04:00
James Shubin
761030b5b8 modules: misc: Make the ssh keygen module more useful 2024-09-26 12:52:27 -04:00
James Shubin
9a752da13d modules: misc: Add a small helper module
Maybe I'll collect enough small snippets that I can keep them in here
until they get split out elsewhere to more appropriate places.
2024-09-26 12:43:26 -04:00
James Shubin
13fc711657 util: Expanding home directory should preserve trailing slash
Because we consider slashes as a directory identifier when needed.
2024-09-26 12:29:43 -04:00
James Shubin
6419f931ee engine: resources: Add a virt-builder resource
This wraps the excellent virt-builder utility which makes setting up new
virtual machines a breeze.
2024-09-26 11:47:40 -04:00
James Shubin
562138cb74 lang: core: os: Add expand_home function
Simple helper that works with ~/foo/ or ~james/foo/ type patterns.
2024-09-25 20:31:18 -04:00
James Shubin
8aac770bcb engine: resources: pkg: Add a small helper for simple installs
Add this utility function for doing single package installs.
2024-09-25 16:26:32 -04:00
James Shubin
80e8c9cadc util: arch: Use small arch util library
Put all of the arch stuff that we can into this library.
2024-09-25 16:25:55 -04:00
James Shubin
87b3dda867 readme: Fixup typo
Reported in GH#775.
2024-09-19 13:22:00 -04:00
James Shubin
b9e093cd6b engine: resources: svc: Reduce unnecessary logging 2024-09-18 21:47:29 -04:00
James Shubin
06a023ca66 engine: resources: sysctl: Be more careful about dir path
More validation is not bad!
2024-09-18 21:39:33 -04:00
James Shubin
ccb4c6244d engine: resources: exec: Improve the docs for a common scenario
I knew this, and now you know it too!
2024-09-18 21:39:03 -04:00
James Shubin
4489e5ce6e engine: graph: autoedge: Quiet down the useless logs
These are so useful, let's silence them.
2024-09-18 21:38:29 -04:00
James Shubin
8df82f0301 docs: Add a new faq entry about deploy.readfile
This may be a common thing people forget.
2024-09-18 21:38:08 -04:00
James Shubin
57b4a7efce lib, engine: graph: Let children directories be readable
We want to be able to put useful scripts in $vardir type places, but if
the perms at the higher levels block this, then that can't work. The
top-level should always be more permissive, and then it grows more
restricted as we descend.
2024-09-18 21:03:58 -04:00
James Shubin
fd508fbc0d engine: resources: Fix typo in svc
It's a typo, right?
2024-09-17 18:10:13 -04:00
James Shubin
a4f368fc9f engine: resources: Add a sysctl resource
Very useful since these are easy to forget!
2024-09-15 23:07:03 -04:00
James Shubin
e7b57a32fd engine: resources: Add a tar resource
This makes tar archives from a list of files and/or directories. It
correctly includes empty directories as well. The code structure is
similar to the gzip resource. While this resource is arguably more
useful than gzip, it was invaluable to write the gzip resource first
since that made writing this one much easier.
2024-09-13 20:04:53 -04:00
James Shubin
06cc63fcb6 util: recwatch: Add a helper function for merging these
I should really rework the recwatch package and API, but I wasn't in the
mood to touch this code today, so this will have to do for now.
2024-09-13 19:48:00 -04:00
James Shubin
e34212a10b engine: resources: gzip: Check unhandled error
This is probably inconsequential, but let's do it since it's not in a
defer.
2024-09-13 19:48:00 -04:00
James Shubin
5f6e07b5e8 engine: resources: gzip: Fix typo 2024-09-13 16:24:01 -04:00
James Shubin
1465c5cdc9 engine: resource: gzip: Remove unneeded waitgroup
I think this was a copy+pasta mistake.
2024-09-13 16:13:03 -04:00
James Shubin
29eebd0d07 lang: core: Move template to golang namespace
I don't think this template function should be in any way authoritative,
so let's namespace it.
2024-09-13 15:51:24 -04:00
James Shubin
5bbc06d8bc engine: resources: Add new gzip resource
This may have lots of uses, particularly for bootstrapping and handoff
if we want to compress payloads. It is also a good model resource for
how to implement such a resource to avoid re-computing the result on
every CheckApply call. Of course if the computation is cheaper than the
hashing of the data this isn't the optimal approach.
2024-09-13 03:32:10 -04:00
James Shubin
9a5f6a5bd3 lang: core, funcs: Use the correct zero type
I wasn't using the correct contained type here.
2024-09-10 23:23:00 -04:00
James Shubin
2e774215e4 lang: core: deploy: Add a function to get the binary path
Useful for bootstrapping new machines.
2024-09-10 23:22:53 -04:00
James Shubin
1327752725 engine: resources: Special log message for unhandled decline
Not sure why a PXE client is sending these... Not sure if it's buggy
firmware or my inability to handle a DHCP corner case.
2024-09-10 23:22:47 -04:00
James Shubin
118f266211 lang: core: local: Add a new vardir function
This gives us a function to return a created vardir folder. It is not
locally namespaced, and a future function will have to namespace one to
each scope.
2024-09-09 18:04:18 -04:00
James Shubin
87a2dfc8f9 engine: local: Add a vardir API to our local API collection 2024-09-09 17:41:12 -04:00
James Shubin
b88ac4603f lang: interfaces Add CallableFunc interface
Add a new interface for callable functions. This will likely be useful
for future versions of the function engine and for the "timeless" work.
2024-09-09 15:49:57 -04:00
James Shubin
28e81bcca3 modules: Add a modules directory for mcl code
Details in the README file.
2024-09-09 15:14:33 -04:00
James Shubin
3d0660559e examples: lang: Add a join example
Good reminder that a lot of the golang stdlib functions are available.
2024-09-09 14:54:45 -04:00
James Shubin
48dc9ad099 test: shell: Disable another flaky test
We need to fix these all eventually, but that day is not today.
2024-09-06 16:24:28 -04:00
James Shubin
fd3a2a1f0f engine: resources: Make consul optional
Licensing has made this non-free. Let's put that behind a build tag for
now, and remove it entirely if no suitable libre replacement is found.
2024-09-03 20:26:38 -04:00
James Shubin
c6e9175e3f engine: resources: Add missing build tag 2024-09-03 20:21:31 -04:00
James Shubin
1a39472734 lang: core: embedded: provisioner: Sometimes this is used
We need better overview of all the PXE/netboot stuff, probably we should
read a spec, but until an expert comes along, we'll have to proceed
incrementally.
2024-09-03 14:48:38 -04:00
James Shubin
bfa88e9b1c engine: resources: Workaround regression in wget2
Apparently wget2 has a serious regression that the HTTP 102 header
throws it off... So let's not send this for now... I'm pretty unhappy
about this, wget used to always be rock solid. Maybe curl deserves a
chance? (This works fine with curl btw.)
2024-08-30 20:33:41 -04:00
James Shubin
a0972c0752 lang, engine: Add a metaparam for catching accidental dollar signs
Let's make our life easier for users!
2024-08-22 20:41:48 -04:00
James Shubin
8dc0d44513 lang: Add an extra fail scenario to our test suite
Let's us write tests for Validate failures.
2024-08-22 20:12:59 -04:00
James Shubin
8594b6e2a9 lang: funcs: Hint the struct_lookup functions better
If we have static information, use it to help unification.
2024-08-21 19:00:51 -04:00
James Shubin
82cac572ca lang: core: fmt: Allow type unification variables for format
This allows some simple cases.
2024-08-21 19:00:50 -04:00
James Shubin
da4f69cd87 lang: ast, core: fmt: Allow unification variables for fmt
This lets us pass through unification variables into the fmt function. I
hope this doesn't break anything, but it's worth trying for now.
2024-08-21 18:52:24 -04:00
James Shubin
e6cb776eb6 lang: ast, core: fmt: Catch invalid nil signatures
We accidentally had a bad error triggered.
2024-08-21 18:50:11 -04:00
James Shubin
7557114b4e lang: ast: Don't send empty ord names for partials
We would accidentally send some empty partials, woops! This reinforces
my belief that we should never pre-allocate list size unless we notice a
performance issue.
2024-08-21 18:00:44 -04:00
James Shubin
001e1a5da0 lang: Remove some error wrapping
Makes errors cleaner to read. The extra context wasn't very helpful.
2024-08-18 19:07:27 -04:00
James Shubin
6f3c3c318b lang: core: Shorten functions with wrapper
This demonstrates how to write a function with the wrapper. Note that
you must not include Init if you're not calling the nested wrapper
function.
2024-08-18 18:29:01 -04:00
James Shubin
654e376be7 lang: core: Add list and map packages
Put the common functionality for those types in there.
2024-08-18 18:28:26 -04:00
James Shubin
211121cdca lang: funcs: Use correct constant 2024-08-18 17:31:58 -04:00
James Shubin
f2d4cac92d docs: Add a short contributing guide
I think this is all common sense, but I thought it might be helpful for
anyone that might not be well-versed with how such projects run.
2024-08-16 23:57:38 -04:00
James Shubin
c5dc9c7650 docs: Add a guide for writing API services
Hopefully this is useful to companies who want to design their services
properly to support modern tooling.
2024-08-16 23:38:27 -04:00
James Shubin
7596f5b572 lang: core: os: Add family functions and variables
Make it easier to do os-specific stuff.
2024-08-07 17:30:15 -04:00
James Shubin
8e9c3b6c1e lang: funcs: vars: Include system package variables in the scope 2024-08-07 17:30:13 -04:00
James Shubin
a93c98402a lang: ast: Add better logging about scope issues
This may help out programmers who aren't sure what's going on.
2024-08-07 17:17:57 -04:00
James Shubin
b04ee4ba22 lang: ast: Pass through the data field for vars 2024-08-07 17:17:57 -04:00
James Shubin
65b104ea55 lang: ast: Split off helpers into util file 2024-08-07 17:17:57 -04:00
James Shubin
562eb643fc engine: resources: Display bytes copied when making a file 2024-08-06 15:24:45 -04:00
James Shubin
80178422db engine: graph, resources: Clean up log messages
The idea is to have a better user experience in the terminal.
2024-08-06 15:12:10 -04:00
James Shubin
e94f39bf2c engine: resources: Cleanups to the svc resource
Some new API's exist that take a context now too!
2024-08-06 14:31:39 -04:00
James Shubin
6c1a33066a engine: resources: The svc resource should reload on notification
Missing feature that is finally landing. I wish this wasn't needed, but
we need fancier plumbing to avoid it.
2024-08-06 14:22:15 -04:00
James Shubin
beca0c3ae6 engine: resources: Plumb through the context and constants
Basically a small cleanup.
2024-08-06 14:13:39 -04:00
James Shubin
7517c83953 engine: resources: Add the systemd service constants
Basically a cleanup to avoid duplicate strings everywhere. This makes it
easier to follow the code too.
2024-08-06 14:02:17 -04:00
James Shubin
0354082f89 engine: resources: Allow symbolic modes for missing files
In 83a747794e a bug was introduced with
the implementation of symbolic modes, that would prevent a file resource
from passing the Validate step if you were using a symbolic mode, and
the file didn't already exist. If you didn't use symbolic modes and
those files weren't absent, then you wouldn't have noticed.

It might be worth looking into the API for symbolic parsing as well.
2024-08-06 13:24:25 -04:00
James Shubin
4abcd9cf01 lang: core: Quiet down the template function by default
We don't need to know this most of the time.
2024-08-01 20:54:55 -04:00
James Shubin
c974820c56 engine: resources: Add log messages for chmod and chown 2024-08-01 20:32:56 -04:00
James Shubin
88670ae7a1 engine: resources: Improve output of log messages
I don't remember ever having this display a pointer address, but it is
now, so let's make this cleaner.
2024-08-01 18:44:44 -04:00
James Shubin
d0ed004b24 examples: lang: Test that each of the mcl examples compiles
We let these rot, so fixup the issues and test them!
2024-07-31 17:29:42 -04:00
James Shubin
6de7d8b254 lang: funcs: Catch non-specific type build error
If you had ambiguous code, and specified an invalid type, this could
sneak through and become a runtime error, instead of a compile-time
error. We fix this and add a test.
2024-07-31 17:29:42 -04:00
James Shubin
bfb5d983c1 lang: types, unification: Don't recurse into private fields
We forgot to omit looking deeper into private struct fields. I don't
know why we didn't catch this earlier, I can only assume some subtlety
changed, since we've previously used many of the resources this would
fail on. Maybe golang broke some API that they didn't consider stable?

This also adds a new test for this, and ensures each resource can be
inspected too!
2024-07-31 17:29:42 -04:00
James Shubin
0a183dfff9 lang: funcs: txn, util: Fix typos 2024-07-25 12:44:58 -04:00
James Shubin
8b54306eb9 examples: lib: Fix these rotted tests
I think the mgmt lib approach is a good idea, even though I'm not
putting much energy into keeping these up to date. Let's at least
re-enable the tests for now, after a few fixups.
2024-07-25 12:39:43 -04:00
James Shubin
fd86b35ce3 docs: Improve the FAQ 2024-07-23 17:26:57 -04:00
James Shubin
d9f8dd53c1 test: Add comment explaining the line length rule issue better 2024-07-23 17:26:18 -04:00
Omar Al-Shuha
ccb0e55d5a examples: lang: Fix env0 example
Change function calls to the correct
one, remove extra argument in getenv
call, and fix typo.
2024-07-08 02:22:57 +02:00
James Shubin
74f747e80b util: password: Fix suspicious dep issue
It seems that without warning, the author of this dep has nuked the old
version, and reorganized the source tree significantly. I'm not an
expert and cryptography routines, but this doesn't make me feel warm
inside. I hope more expert researchers could look into this so that we
avoid supply chain attacks.
2024-07-07 12:47:14 -04:00
James Shubin
aa03b5ce2f lang: core: iter: Add filter iterator function
This was fun to write and adds a new core iterator function.
2024-07-03 21:25:19 -04:00
James Shubin
e747e12002 examples: lang: Fixup a few examples
We might change unification to allow naked single strings with fancier
unification, but let's leave it as is for now and see how often it comes
up.
2024-07-02 23:50:24 -04:00
James Shubin
d1753c592a lang: core: iter: Misc formatting fixes
Also fix up the examples.
2024-07-02 23:49:56 -04:00
James Shubin
7a35bef7ac test: Improve comment parser to skip code blocks
It might be nicer to have some code blocks all by themselves on a single
line.
2024-07-02 13:24:55 -04:00
James Shubin
e10e92596f lang: types: Add stringer information manually
This lets us get the more correct lowercase versions of type kinds in
error messages. (These match what the user would type.)
2024-07-01 18:35:20 -04:00
James Shubin
28253c4bd2 lang: Move stateful test objects into a per-test mode
Was this causing failures? Does this make things much slower?
2024-07-01 18:34:42 -04:00
James Shubin
f2976deb02 pgraph, lang: ast: Fix failing tests due to non-deterministic topo sort
This causes inconsistent type unification when running our tests. It's a
bad user experience too.
2024-07-01 18:34:24 -04:00
James Shubin
14577a0c46 lang: Add modern type unification implementation
This adds a modern type unification algorithm, which drastically
improves performance, particularly for bigger programs.

This required a change to the AST to add TypeCheck methods (for Stmt)
and Infer/Check methods (for Expr). This also changed how the functions
express their invariants, and as a result this was changed as well.

This greatly improves the way we express these invariants, and as a
result it makes adding new polymorphic functions significantly easier.

This also makes error output for the user a lot better in pretty much
all scenarios.

The one downside of this patch is that a good chunk of it is merged in
this giant single commit since it was hard to do it step-wise. That's
not the end of the world.

This couldn't be done without the guidance of Sam who helped me in
explaining, debugging, and writing all the sneaky algorithmic parts and
much more. Thanks again Sam!

Co-authored-by: Samuel Gélineau <gelisam@gmail.com>
2024-07-01 18:33:47 -04:00
James Shubin
4e18c9c67a lang: Plumb through the unified state facility 2024-07-01 16:07:14 -04:00
James Shubin
d326917432 lang: interfaces: Add some unification basics
This includes the GenericCheck helper which we'll use everywhere, and
the standard single invariant which we use throughout.
2024-07-01 16:07:14 -04:00
James Shubin
ad4eb86262 lang: unification: util: Add the core unification helpers
This adds the core unification helper functions that do the core work of
solving the invariants. This includes the actual Unify, OccursCheck, and
Extract which is sometimes known as "zonk".

A few other small functions are also included.

Co-authored-by: Samuel Gélineau <gelisam@gmail.com>
2024-07-01 16:06:58 -04:00
James Shubin
5c73e7c582 lang: types: Add a facility for printing consistent unification vars
When we look at unification variables from two different places, the
default printer will always start numbering them from ?1 and therefore
if we look at two unrelated systems, they might both print as ?1 when
they are in fact different pointers.

We don't collect them all by default since it's usually not necessary
except for debugging, but in those situations, we want a consistent
unification store which we can pass around to get sensible debug output.
2024-07-01 14:16:11 -04:00
James Shubin
dc33d9aab7 lang: types: Add a comparable helper to our types library 2024-07-01 14:14:20 -04:00
James Shubin
cdc2439f89 lang: types: Add an iterator helper to our types lib 2024-07-01 14:13:41 -04:00
James Shubin
318ee0d002 lang: types: Fixup small log messages 2024-07-01 14:12:22 -04:00
James Shubin
653299a88f lang: types: Plumb in a unification variable into our type
This is used for representing a unification variable in our type during
type unification. For example, this allows us to have a [?1] or a
map{?1:[?2]} and so on...
2024-07-01 14:11:03 -04:00
James Shubin
6066cbf075 util: Add disjoint package to implement a union find datastructure
This is a fascinating, and incredibly simple data structure. I hope I
can end up using it for more than just type unification!

Thanks to Sam who taught me about its existence.
2024-07-01 14:05:48 -04:00
James Shubin
2b3a41fefa pgraph: Fix rare panic if Sprint is badly used
Be nicer.
2024-07-01 14:05:48 -04:00
James Shubin
5ca9f7fa38 test: Skip link check by default
We should run this periodically or put it in a separate job, as it's
causing the tests to fail all the time. I expect those sites may be
blocking github as they see it as a DOS.
2024-07-01 13:40:14 -04:00
xlai89
201cf091d5 test: Add a links checker and fix some links 2024-06-17 14:17:39 -04:00
Cian Yong Leow
09e53bfd3f engine: resources: file: Remove Validate owner/group Checks
The owner/group of a file should not be validated on the host until runtime. This removes the checks in Validate() that were happening before the execution of the resource graph (and therefore bound to fail if the system was being bootstrapped).
2024-06-17 13:52:04 +01:00
James Shubin
3c661ab674 lang: core: embedded: provisioner: Version flag needs a unique name
Conflicts with the stock --version.
2024-05-23 07:57:30 -04:00
James Shubin
415e22abe2 lang: core, funcs, types: Add ctx to simple func
Plumb through the standard context.Context so that a function can be
cancelled if someone requests this. It makes it less awkward to write
simple functions that might depend on io or network access.
2024-05-09 19:25:46 -04:00
James Shubin
3b754d5324 docs: Fix markdown failing
It fails locally, but not in CI, and I don't know why.
2024-05-05 15:34:04 -04:00
James Shubin
7a568627e9 docs: Update dead links 2024-05-05 15:34:00 -04:00
James Shubin
328360eea8 docs: Add addition to style guide for pointer receivers 2024-04-28 16:14:05 -04:00
James Shubin
7ae3ba4483 mod: Run go mod commands
This was done with go get -u ./... followed by go mod tidy.
2024-04-27 13:31:32 -04:00
Joe Groocock
351a61c0cd engine: resources: docker: Update docker
Several types were renamed and moved

Signed-off-by: Joe Groocock <me@frebib.net>
2024-04-27 10:52:29 +00:00
James Shubin
c12452b3ce misc: Move to golang 1.21
Unfortunately, this also breaks go-mod-upgrade with:

upgrade failed error=Error running go command to discover modules: exit
status 1 stderr=go: loading module retractions for
golang.org/x/mod@v0.16.0: version "v0.17.0" invalid: resolves to version
v0.17.1-0.20240315155916-aa51b25a4485 (v0.17.0 is not a tag) go: loading
module retractions for golang.org/x/sync@v0.6.0: version "v0.7.0"
invalid: resolves to version v0.7.1-0.20240304172602-14be23e5b48b
(v0.7.0 is not a tag)
2024-04-25 13:01:41 -04:00
James Shubin
0e92d190cc make: Add easy error message for common issue
This can happen if the golang tools are angry. Make it easier for the
user to debug and fix.
2024-04-25 12:38:50 -04:00
James Shubin
453cd4409e lang: ast: Remove unnecessary metaparam exclusive
Originally, I considered having more than one way to express the meta
param. After thinking about it for longer, it probably makes sense to
have a second meta param if necessary, and to avoid the exclusive.
2024-04-18 00:44:34 -04:00
James Shubin
51cf1e2921 lang: ast: The res and edge names should not use exclusives
This removes the exclusive from the res names and edge names. We now
require that the names should be lists of strings, however they can
still be single strings if that can be determined statically.
Programmers should explicitly wrap their variables in a string by
interpolation to force this, or in square brackets to force a list. The
former is generally preferable because it generates a small function
graph since it doesn't need to build a list.
2024-04-18 00:07:53 -04:00
James Shubin
dc45c90ccd lang: Add common type to global variables
We use the list of strings so often, we might as well give it a global
variable.
2024-04-16 14:31:03 -04:00
James Shubin
6782d65577 test: shell, lang: core: embedded: provisioner: Check it compiles
Add a test to guarantee we continue to keep compiling, in case something
in the language changes.
2024-04-16 14:30:56 -04:00
James Shubin
68ee163eb1 entry, lang: core: embedded: provisioner: Allow more than one entry
This changes the entry API slightly to allow for more than one entry
registered, which makes building, testing and user tooling easier.
2024-04-16 14:11:34 -04:00
James Shubin
bc4b5d96b0 lang: core: embedded: provisioner: Better name for firewall entries 2024-04-15 15:03:27 -04:00
James Shubin
909dbb531d lang: ast: Fix small typos 2024-04-06 15:32:31 -04:00
Felix Frank
a2654bdc69 test: Send error messages to stderr, where they belong
When error messages are written to stdout, they will be considered as
output in case we want to fail from inside $( ) or backticks, and then
the error does not end up on the terminal.
2024-04-02 21:32:05 -04:00
Felix Frank
edcb04d1a9 misc, test: Some quality of life improvements
Add a fold in github actions output around the ragel build.

Run the commit-message test locally, so that error can be detected
before pushing to CI. We also now accept two-letter topics.

Some minor improvement in the testing scripts.
2024-04-02 21:15:51 -04:00
Felix Frank
29ec867ac7 gapi: Bring back puppet and langpuppet
This reverts commit e767655ede.

In addition, it applies required changes to function with the new CLI backend.
2024-04-02 21:07:02 -04:00
Julian Rüth
22873b3c3f docs: Fix docker build instructions
fixes #752
2024-04-03 01:01:47 +03:00
Julian Rüth
ede5db18d7 docs: Binaries are not outdated currently 2024-04-03 01:01:37 +03:00
James Shubin
964b1dc58a docs: Add release notes for 0.0.26
I send these out by email and then archive a copy here. Sign up to the
mgmt partner program for early access. Ping me for details.
2024-03-30 19:03:31 -04:00
James Shubin
1b00af6926 lang: Add unification optimizations
This adds a unification optimizations API, and uses it to optimize the
embedded provisioner. With these turned on, type unification drops from
around 1m45s to 2.5s which is a 40x speedup.
2024-03-30 18:05:16 -04:00
James Shubin
ddf1be653e lang: Plumb data and unification strategy through the lang struct
This adds some plumbing to pass values into the lang struct.
2024-03-30 17:36:49 -04:00
James Shubin
cede7e5ac0 lang: Structurally refactor type unification
This will make it easier to add new solvers and also cleans up some
pending issues.
2024-03-30 16:55:20 -04:00
James Shubin
964bd8ba61 util: Skip badly numbered versions
This let's us skip over pre-release versions like "40 Beta" or similar.
2024-03-30 13:54:09 -04:00
James Shubin
a1db219fd2 docs: Add release notes for 0.0.25
I send these out by email and then archive a copy here. Sign up to the
mgmt partner program for early access. Ping me for details.
2024-03-27 17:21:08 -04:00
James Shubin
241be1801b make: Update remote path 2024-03-27 00:25:23 -04:00
James Shubin
4d9c78003a lang: core: embedded: Add standalone provisioning tool
This commit adds the ability to build a standalone provisioning tool.
This is the first useful public mcl code base as well. It is not
perfect, but does serve as a rough starting point to show what is
possible. In the future as the language and the engine evolve, this will
likely get more elegant, and also grow new features.

To build this, run `make clean && GOTAGS='embedded_provisioner' make`.
To run this, run `mgmt provisioner`.
2024-03-26 20:33:43 -04:00
James Shubin
375fe19f52 make: Build some extravagant release hacks
This should make a maintainers life easier. (I hope!) The .release files
contain the "magic name" for the respective release type. If they start
with a # then they are skipped. You shouldn't need to keep bumping the
distro versions in the Makefile anymore.
2024-03-26 20:23:06 -04:00
James Shubin
1895c63e89 make: Clarify how mkdir runs 2024-03-26 18:40:44 -04:00
James Shubin
33a00436b1 make: Selectively include files if they exist
Make our Makefile fancier...
2024-03-26 17:56:35 -04:00
James Shubin
8ad232d96a make: Update fedora release versions
These were untested.
2024-03-26 17:07:39 -04:00
James Shubin
01b7721b13 make: Add release targets for standalone binary builds
Many people might just want to download a single binary.
2024-03-26 16:57:06 -04:00
James Shubin
46b2fe0eba docs: Add old release notes
These release notes used to live on the mailing list at:
https://listman.redhat.com/archives/mgmtconfig-list/ until Red Hat
killed off this excellent service recently.

I'm adding them all here for reference.

Before 0.0.9 there were no release notes.
2024-03-26 14:38:26 -04:00
James Shubin
89784e86bd lang: gapi, unification: Shutdown unification quickly when asked
This is a bit of a hack until we improve the GAPI a bit, but will let us
shut down type unification a bit faster if we want to interrupt a long
running operation. Hopefully our future algorthmic performance
improvements will obliviate the need for this to be a common issue.
2024-03-22 03:23:46 -04:00
James Shubin
c4b14ac40d lang: download, cli: Move to new git version
Apparently there were some security issues with the old one.
2024-03-22 02:56:59 -04:00
James Shubin
12b519a543 lang: gapi: Improve logging to be cleaner 2024-03-22 02:56:59 -04:00
James Shubin
f75f6623b3 util: recwatch: Improve godoc 2024-03-22 02:56:59 -04:00
James Shubin
43b5a0ea6a util: recwatch: Remove last use of core logger
We've cleaned up quite a bit, yay!
2024-03-22 02:56:59 -04:00
James Shubin
3541954df8 lib, pgp: Remove direct logger usage
This cleans things up a little bit more.
2024-03-22 02:56:59 -04:00
James Shubin
e0d6d35b56 entry: Execute embedded commands if you use their correct name
I think this is the more intuitive way to do things. Force the user the
explicitly choose the embedded command which prevents accidental running
of a different command, and allows us to potentially have multiple
commands in the future.
2024-03-22 01:36:04 -04:00
James Shubin
8cf7719476 cli: Refactor the logger setup 2024-03-22 01:36:04 -04:00
James Shubin
d2188609e4 entry: Add a new API for log related things
This needs improvement to eventually let the embedded application
disable logs from the main execution, or at least filter them through
itself to decide what to keep. That's a future complex project though.
2024-03-22 01:13:21 -04:00
James Shubin
3e180eafb4 cli: Refactor out log to the top level
Remove some cruft at the same time.
2024-03-22 00:55:03 -04:00
James Shubin
b8a3c39984 util: password: Add a new helper util function for passwords
Also fix some doc issues from an earlier copy paste.
2024-03-22 00:55:03 -04:00
James Shubin
946468dc99 lang: ast: Ensure a list doesn't sneak through type interpolation
If we had a single list wrapped in an interpolated string, it could
sneak through type unification, which is not correct. Wrapping a
variable by interpolation in a string, must force it to be a string.
2024-03-20 18:36:47 -04:00
James Shubin
340a832884 lang: ast: Optimize unification invariants for common case
This optimizes the list of type unification invariants that we generate
for the common case where a resource or edge name is known statically.

For one code base this halved the type unification time in half. More
work can be done though!
2024-03-20 18:07:59 -04:00
James Shubin
388d08e245 engine: resources: Add a dhcp range resource
This adds the ability to offer a dhcp lease to someone when we don't
know their mac address in advance.

This also uses the extended autogrouping API to keep the internal API
simpler.
2024-03-20 17:45:06 -04:00
James Shubin
8a78907977 engine: autogroup, traits, graph: Extend autogroup API
This extends the autogrouping API so that a child can easily get a
reference to the parent that it is autogrouped in. This can simplify the
API for some resources when it makes sense to allow them access to the
parent handle. Use sparingly and intelligently!
2024-03-20 17:45:06 -04:00
James Shubin
6347e275d3 util: Add a fedora mirror lookup function 2024-03-18 15:38:57 -04:00
James Shubin
c8f19e0d96 mod: Ran a combination of go get, go mod tidy, and go-mod-upgrade
These tools are all junk. Maybe one day I can hire someone to fix things
the right way.
2024-03-18 15:37:22 -04:00
James Shubin
ba6d816186 lang: unification: Clean up the solver plumbing
This refactors the solver into a separate struct that can be extended as
needed.
2024-03-16 01:38:33 -04:00
James Shubin
849de648f0 engine: resources: Add a creates field for exec
This adds a standard gate that prevents execution if a file exists. Of
note, this also adds a watch on it, so we can have a proper watched exec
resource without a watch cmd.
2024-03-16 01:21:32 -04:00
James Shubin
4d18044851 util: Move recwatch package to util directory
This makes more sense and this is also a reminded that it needs a bit of
a cleanup.
2024-03-16 01:21:32 -04:00
James Shubin
d6a6734b65 lang: Plumb the ctx deeper into the lang API 2024-03-16 01:21:32 -04:00
James Shubin
10319dd641 lang: Plumb through a context into unification
If we have a long type unification, we might want to cancel it early.
This also helps us visualize where we want context to be seen.
2024-03-16 01:21:32 -04:00
James Shubin
a8b945e36e lang: types, core: Skip over invalid functions
It's not clear how best to support complex functions with struct fields
in templates at the moment. Skip these for now.
2024-03-16 01:21:32 -04:00
James Shubin
719c56e754 lang: core: convert: Add a function to convert an int to a str
Faster than using printf.
2024-03-16 01:21:32 -04:00
James Shubin
cbf10bdb44 lang: core: os: Add a simple distro uid parsing function
I had previous good success using this kind of pattern for discussing
distro UID's. Let's put this in as a reminded to see if it's worth
pursuing longer term.
2024-03-16 01:21:32 -04:00
James Shubin
ee60c428b2 engine: resources: Clarify log message
This way you don't have to remember which way it works.
2024-03-16 01:21:32 -04:00
James Shubin
a410b4981f lang: Add timing information to lang execution
This makes performance optimizing a little easier.
2024-03-16 01:21:32 -04:00
James Shubin
f973009b83 lang: Rephrase the scope building log messages 2024-03-16 00:36:06 -04:00
James Shubin
e0ace35525 engine: resources: The http proxy should support streaming files
This adds basic support for streaming files directly from the download
server. This avoids clients timing out if they are blocked while first
waiting for a giant file to download.
2024-03-16 00:36:06 -04:00
James Shubin
cf49d9f784 engine: resources: Improve proxy conventions
The earlier path mangling code was incorrect. I've taken more time to
understand the correct use case and I've improved it. I've also split
out the parser logic and added tests, so this should either stay stable
or grow new tests and fixes if we find new issues.
2024-03-16 00:36:06 -04:00
James Shubin
79d5873445 engine: resources: Also look at stderr
We might have an executing program which prints on stderr which we were
effectively ignoring. Read from that too.
2024-03-16 00:36:06 -04:00
James Shubin
c5e3e0ee70 engine: resources: Add firewalld resource
This is a simple firewalld resource to make the seamless opening of
firewall ports in standalone laptop (and other) environments easy.
2024-03-16 00:36:06 -04:00
James Shubin
6976f5f3f0 util: Add simple distro release version function
This helps us determine the latest version automatically as a user
convenience.
2024-03-08 17:57:13 -05:00
James Shubin
1f37ac83e4 lang: types: Fix capitalization 2024-03-08 17:57:13 -05:00
James Shubin
36ebddf986 util: Add a terrible heuristic for getting ethernet devices
I don't know of a better way. This will have to do for now.
2024-03-08 17:57:13 -05:00
James Shubin
b496f8d70a util: password: Add a helper to generate salted, hashed passwords
I have no idea if I'm doing this correctly, and I have no idea if the
library I am using is doing this correctly. Please check!
2024-03-08 17:57:13 -05:00
James Shubin
9351eee3f1 lang: funcs: simple: Add a function from struct generator
This is a helper function that can generate a bunch of functions from a
struct type. This is most useful when using a CLI args struct for
command line parsing and then storing the values as functions.

An alternative version of this might choose to return all of the values
as a single giant struct.
2024-03-08 17:57:13 -05:00
James Shubin
d412d6502d lang: gapi: Fix a very rare nil pointer issue
This only showed up while debugging a bug I caused, but technically if
we fail in Init() then we could hit this nil pointer condition.
2024-03-08 17:57:13 -05:00
James Shubin
b85f81d529 lang: types: Add simple parsing of net.HardwareAddr
If we want to use special struct types from our CLI parser, we also need
to be able to both identify, and convert them to our language type and
value representations.

For as long as we don't have fancier types in our language, these should
both be strings. Tests and extensions to these additions are welcome!
2024-03-08 17:57:13 -05:00
James Shubin
4140492d56 lang: ast: Embedded imports have function scopes too
If we're importing an embedded module, we need to also include any
possible system scope imports (like functions) that might be using the
same namespace. These might be used by a custom CLI frontend to extend
the code with values from the CLI parsing, for example.
2024-03-08 17:41:57 -05:00
James Shubin
59e133d3bc make: Find on all mcl files
We forgot to port this line when we moved the core package up one
directory.
2024-03-08 17:41:57 -05:00
James Shubin
90f6d4e563 legal: Update http to https 2024-03-05 01:05:50 -05:00
James Shubin
3e31ee9455 legal: Additional permission under GNU GPL version 3 section 7
With the recent merging of embedded package imports and the entry CLI
package, it is now possible for users to build in mcl code into a single
binary. This additional permission makes it explicitly clear that this
is permitted to make it easier for those users. The condition is phrased
so that the terms can be "patched" by the original author if it's
necessary for the project. For example, if the name of the language
(mcl) changes, has a differently named new version, someone finds a
phrasing improvement or a legal loophole, or for some other
reasonable circumstance. Now go write some beautiful embedded tools!
2024-03-05 01:04:09 -05:00
James Shubin
d52c90ede4 cli, entry, lang: Add an entry package for embeddable CLI's
This adds a new entry package that allows embedded programs to exist
inside of mgmt. This took a lot of refactoring to get the API right, but
I think it's incredibly elegant now. There is a chance we tweak things a
bit, but it's a good first start. All-in-one programs are coming soon!
2024-03-05 01:03:54 -05:00
James Shubin
9527d0dcbd lib, cli: Move cli struct tags and embed this datastructure
This moves over the cli `arg` struct tags which are used to generate and
parse things on the command line. Furthermore, we then embed this data
directly in our more general parser struct so that we avoid duplication.
Finally, since the data shares a common struct type, we don't need to do
the manual field-by-field copying to pull things in!
2024-03-03 20:07:04 -05:00
James Shubin
51d21b8dab lib: Split out all of our config data as a separate struct
Hopefully this will make things more reusable.
2024-03-03 19:18:32 -05:00
James Shubin
601fcf40c4 cli, gapi: empty, lang, yaml: Refactor the args structs
Put these datastructures into an external package so they can be re-used
for parsing elsewhere.

Since we remove these dependencies, we need to manually import the
GAPI's so that they register. Despite efforts to embed them deeper into
the import tree without cycles, this failed. Logically what this told me
is that it actually makes sense to allow a different binary with only
one of the multiple GAPI's contained within.
2024-03-03 17:17:21 -05:00
James Shubin
d537c3d523 cli: Lookup subcommand names dynamically
This lets us get the chosen subcommands dynamically and removes the
eventual need for three package imports.
2024-03-03 15:50:54 -05:00
James Shubin
a65c87b584 gapi: Add a new Names API
This gets the list more directly. We could also just lookup a key in the
big map to get the same effect, but this is more logical for now. This
is useful for removing the importing of three packages.
2024-03-03 15:50:54 -05:00
James Shubin
589a5f9aeb cli, lib, lang: Port to new cli library
The new version of the urfave/cli library is moving to generics, and
it's completely unclear to me why this is an improvement. Their new API
is very complicated to understand, which for me, defeats the purpose of
golang.

In parallel, I needed to do some upcoming cli API refactoring, so this
was a good time to look into new libraries. After a review of the
landscape, I found the alexflint/go-arg library which has a delightfully
elegant API. It does have a few rough edges, but it's otherwise very
usable, and I think it would be straightforward to add features and fix
issues.

Thanks Alex!
2024-03-01 21:02:55 -05:00
James Shubin
e767655ede cli, puppet, langpuppet: Remove puppet GAPI's
I'm currently refactoring the CLI code. Unfortunately this means a
pretty big churn in the various GAPI frontends. Since nobody is actively
using the puppet frontend code, I'm removing it for now. If someone is
actively using it, and wants to either port it to the new API, or
sponsor the porting of it to the new API, I'm happy to allow it back in.

Sorry Felix, it was a fun idea, and I loved seeing it work, but I can't
personally afford the maintenance cost of having this in right now.
2024-03-01 21:02:55 -05:00
James Shubin
62295e370c test: shell: Disable flaky test
Not sure what's up, but we should investigate eventually.
2024-03-01 21:02:55 -05:00
James Shubin
f818f5ccf5 util: errwrap: Add more tests 2024-02-28 16:37:28 -05:00
James Shubin
f28d22d20f cli: Typo fix 2024-02-28 16:36:49 -05:00
James Shubin
80af171a35 cli: Further refactor our input cli data 2024-02-28 16:36:49 -05:00
James Shubin
71c54ab212 cli: Refactor even more code out of cli package
Quite honestly, I'm trying to clean things up, and removing anything
non-consequential will help!
2024-02-28 16:36:49 -05:00
James Shubin
c37ff3efce lang, test: Fix capitalization for consistency 2024-02-28 16:36:49 -05:00
James Shubin
dd0e67540f all: Remove deprecated io/ioutil package
Porting everything to the newer imports was trivial except for one
instance which required a very small refactor.
2024-02-28 16:36:49 -05:00
James Shubin
8db41e7701 cli: Rename confusing obj variable name 2024-02-28 16:36:49 -05:00
James Shubin
37569aae17 lang: core: Add package documentation string 2024-02-28 16:36:49 -05:00
James Shubin
ffd6385dd5 lang: inputs: Add visualization of parsed input struct
Occasionally helps with debugging.
2024-02-28 16:36:49 -05:00
James Shubin
abe3e0e7a5 lib: Split off the CLI portions into a separate package
This cleans things up a bit more and forces the lib package to not
contain any accidental CLI parsing code.
2024-02-28 16:36:49 -05:00
James Shubin
70b5ed7067 lang: Add an embedded package for embedded imports
This adds a new "embedded" package which can be used to import
system-like packages that are embedded into the binary.
2024-02-28 16:36:49 -05:00
James Shubin
9d8beb85d7 lang: ast: Refactor import scope for sanity
This makes it easier to integrate future additions to the import code.
2024-02-28 16:36:49 -05:00
James Shubin
296fc484ba test: shell: Disable flaky test
Legacy yaml anyways...
2024-02-22 17:41:13 -05:00
James Shubin
ad900fc8f1 etcd: Store the scheme as a constant 2024-02-22 17:41:13 -05:00
James Shubin
f60c25aacf util: Add afero relpath implementation
This new filesystem implements a relative filesystem modifier which can
be useful for converting between absolute filesystems rooted at / and
relative ones. This is particularly useful when interfacing with the
golang embed package.

The upstream requires a CLA, so we'll just store this here instead.
https://github.com/spf13/afero/pull/417
2024-02-22 17:41:13 -05:00
James Shubin
d6cf595899 lang: Unnested the core package from the functions dir
The core package could contain non-functions, so we might as well move
it upwards.
2024-02-22 17:19:02 -05:00
James Shubin
9329ed1e37 lang: ast: Add a new flavour and signature for scope importing
We'll need this more flexible version for the future.
2024-02-22 13:46:27 -05:00
James Shubin
90628dc5c1 engine: Improve docs
Things were messy here, give this a small cleanup.
2024-02-22 13:46:27 -05:00
James Shubin
73ae197d20 util: Add scheme and path support to our fs util struct
This lets us have a custom URI when wrapping an Afero.Fs interface.
2024-02-22 13:46:27 -05:00
James Shubin
15fa6b82a5 lang: funcs: core: Formatting cleanup 2024-02-22 13:36:46 -05:00
James Shubin
d887e7fea5 util: More error handling 2024-02-22 13:36:46 -05:00
James Shubin
871f0e73c0 lang, lib, util: Rename fs to be more unique
Trying to do a big refactor and this will help.
2024-02-22 13:36:46 -05:00
James Shubin
d117cb8ed5 util: Improve tree printing function
This makes it behave more like the core GNU tree util.
2024-02-22 13:36:46 -05:00
James Shubin
733d7fb55f lang: ast: Clean up code slightly 2024-02-22 13:15:39 -05:00
James Shubin
6ae3481ae9 engine, lang, gapi: Split out some functions to a writeable API
Start breaking down the filesystem interface to make things more
flexible.
2024-02-22 13:15:39 -05:00
James Shubin
b7efd94147 lang: Pass through the fs and be consistent in usage
This simplifies the API by passing through the filesystem so that
function signatures don't need to be as complicated, and furthermore use
that consistently throughout.
2024-02-22 13:15:39 -05:00
James Shubin
a05b6a927e lang: funcs: core: os: Add an args function to read argv 2024-02-22 13:15:39 -05:00
James Shubin
30648a7858 lib: Print out timing from core graph generation
This should help us determine what steps need improving. As it turns
out, autoedges are approximately the slowest. (Highly dependent on the
specific mcl code being used.)

The trend is clear though; note the units:

main: new graph took: 913.653µs
main: auto edges took: 9.273807153s
main: auto grouping took: 28.690819ms
main: send/recv building took: 566ns

main: new graph took: 779.255µs
main: auto edges took: 4.03670168s
main: auto grouping took: 37.682101ms
main: send/recv building took: 121.017µs

main: new graph took: 1.157479ms
main: auto edges took: 3.794132165s
main: auto grouping took: 49.732836ms
main: send/recv building took: 95.921µs

main: new graph took: 900.937µs
main: auto edges took: 7.206085s
main: auto grouping took: 25.508671ms
main: send/recv building took: 489ns

main: new graph took: 794.224µs
main: auto edges took: 4.313729756s
main: auto grouping took: 47.970533ms
main: send/recv building took: 207.62µs

main: new graph took: 884.49µs
main: auto edges took: 7.585529786s
main: auto grouping took: 24.327938ms
main: send/recv building took: 72.741µs

main: new graph took: 774.157µs
main: auto edges took: 2.827380129s
main: auto grouping took: 28.303023ms
main: send/recv building took: 85.246µs

main: new graph took: 746.841µs
main: auto edges took: 2.775868117s
main: auto grouping took: 33.11291ms
main: send/recv building took: 104.875µs

main: new graph took: 796.445µs
main: auto edges took: 2.71556122s
main: auto grouping took: 24.03827ms
main: send/recv building took: 106.414µs

main: new graph took: 1.217452ms
main: auto edges took: 2.908416104s
main: auto grouping took: 61.175916ms
main: send/recv building took: 92.328µs

main: new graph took: 807.894µs
main: auto edges took: 3.222089261s
main: auto grouping took: 40.032629ms
main: send/recv building took: 106.49µs

main: new graph took: 986.963µs
main: auto edges took: 3.538425263s
main: auto grouping took: 30.660849ms
main: send/recv building took: 99.74µs
2024-02-22 13:15:39 -05:00
James Shubin
26bce5ddc0 engine: resources: Add a new http:proxy resource
This adds a new caching http proxy resource that can be autogrouped into
the core http:server resource, and which caches and serves files that it
received from a different http server. It first pulls them down when
they are initially requested, which makes it possible for us to use this
for provisioning a Linux installation without having to pre-rsync the
entire package repository.
2024-02-22 13:15:39 -05:00
James Shubin
bd708159a1 engine: resources: Close files after use
Don't need to wait for gc.
2024-02-22 13:15:39 -05:00
James Shubin
87ce637bd2 engine: resources: Add a helper function for creating errors 2024-02-22 13:15:39 -05:00
James Shubin
7698b1b5fd engine: resources: Add an http error wrapper
This makes it easier to return errors more succinctly.
2024-02-16 08:06:20 -05:00
James Shubin
e256d886e0 engine: resources: Add a fancy http error wrapper
This lets use use a regular error signature but also specify which http
error code we want to return.
2024-02-16 08:06:20 -05:00
James Shubin
6f268e3a40 engine: resources: Fix up docstring
This was a copy+pasta mistake.
2024-02-16 08:06:20 -05:00
James Shubin
57910470a9 docs: Add new talks from Belgium 2024 2024-02-09 10:49:35 -05:00
James Shubin
890b6e9a28 docs: Add empty list declarations to style guide
This seems to have come up lately, so add my reasoning. Debate is
welcome-- to a point.
2024-02-09 04:48:16 -05:00
Eng Zer Jun
b09b21e939 lang: ast: Remove redundant nil check
From the Go specification [1]:

  "1. ... For a nil slice, the number of iterations is 0."

Therefore, an additional nil check for around the loop is unnecessary.

[1]: https://go.dev/ref/spec#For_range

Signed-off-by: Eng Zer Jun <engzerjun@gmail.com>
2024-02-09 04:31:00 +08:00
James Shubin
edf47a1737 lang: gapi, docs: Add a --skip-unify option to speed up run
This is useful for run and dangerous for deploy. If we make type
unification blazing fast, we should probably get rid of this option.
2024-02-08 10:49:22 -05:00
James Shubin
5b8a1ce821 readme: Update links 2024-02-02 11:12:49 -05:00
Oliver Lowe
d47869ac2f lang: types: Fail build if stringer run from GOROOT
If stringer is run from GOROOT, it can't find any packages; same error
as in https://go.dev/issue/31843. And since GOCACHE is always ignored
(see issue above) we can have a bare go generate command.
2024-02-01 19:49:01 -05:00
James Shubin
8ea7d4cf84 engine: graph, util: Restore send/recv behaviour
A regression in 4b0cdf9123 caused the
basic send/recv functionality to break for simple scenarios. This was
due to inadequate testing, and a partial misunderstanding of the
situation.

New testing should hopefully catch more cases, but send/recv and
compile-time checks are still not as complete as is probably possible.
2024-01-31 22:59:58 -05:00
James Shubin
fc49888ba2 examples: lang: Add a simpler map-iterator example 2024-01-31 16:02:47 -05:00
James Shubin
d0d6a4d1a1 engine: resources: Log some tftp errors
This was helpful in debugging.
2024-01-31 13:09:58 -05:00
James Shubin
961370c43c engine: resources: Log dhcp mac addresses
Useful for identifying who is trying to connect when they're not in the
system yet.
2024-01-31 13:09:58 -05:00
James Shubin
902f4c957a lang: Print stats for debugging if function engine takes too long
If we don't startup fast enough, print some debugging information. We
should eventually change this to print the list of functions that aren't
started yet, and also to give each function entry a better String method
so that we have a better idea of what everything is.
2024-01-28 23:16:25 -05:00
James Shubin
7e1a4dea6c engine: resources: The http:file resource should allow directories
This expands the use of the http:file resource to allow it to be used as
a directory root.
2024-01-28 23:16:25 -05:00
James Shubin
c78ef29bda util: safepath: Add AbsPath and RelPath
These add some new types for when you don't know if something is a file
or a directory, but you do know if it's absolute or relative.
2024-01-28 23:16:20 -05:00
James Shubin
305a4ab6dd examples: lang: Move to v3 of tftp client 2024-01-28 15:55:24 -05:00
James Shubin
7777107d83 lang: funcs, interfaces: Add the Graph method
The Txn wants to be able to call graph directly.
2024-01-24 20:49:17 -05:00
James Shubin
67c72a0e86 engine: resources: Add an http:flag resource
This adds a new http:flag resource which can autogroup into an
http:server resource to receive actions from client HTTP requests, and
forward these values on to other resources.
2024-01-22 20:37:53 -05:00
James Shubin
e7a89a4a42 engine: Resource Cmp should be relaxed
When graph swapping (which is quite common) we only use the newly-made
resource if the Cmp function between the two shows a difference. If the
old resource has previously received a value via send/recv, then when it
is compared to the new value, it will almost always be different. As a
result, we need to run send/recv on the newly made graph to make sure it
has up-to-date values before we compare. This has to happen after
autogrouping since the resources can often be autogrouped and any child
grouped resource will cause a remake of all of the other children and
parents.

It turns out that the actual send/recv properties were being compared as
well, and for unknown reasons (tunnel vision perhaps) they are often not
identical. Skip comparing these for now until we find a fix or
understanding of how to make them identical.
2024-01-22 20:37:53 -05:00
James Shubin
6761984f2c lib: Refresh new graph with received values from previous graph
This pulls in the Send/Recv values from the previous graph so that our
Cmp functions are more likely to not remake resources that should
otherwise not have changed. Unnecessary remakes can destroy the private
state of a resource which can make certain operations impossible.
2024-01-22 20:37:53 -05:00
James Shubin
1630eafbe3 engine: graph: Allow Recv overriding for Send/Recv
This allows us to pass in an alternate implementation for Recv, if we
want to temporarily use a different data source.
2024-01-22 20:37:53 -05:00
James Shubin
091bf3a64c engine: graph: Add a function for printing Send/Recv logs 2024-01-22 20:37:53 -05:00
James Shubin
e5a189b8c6 engine: graph: Split SendRecv off from the engine
It can be used in more places if it's not tied to the engine struct.
This also changes the signature so that more information is returned.
This can be used for logging or other useful things. Of note, this
happens to be the same struct as already exists. It's used for
convenience since it happens to match up! Of course they're related.
2024-01-22 20:37:53 -05:00
James Shubin
f68b34a485 engine: Add a mapper function for comparing two res graphs
This gives us a simple mapping between a new resource and an old one. We
compare by kind and name because those two values are our uniqueness
constraint in the resource graphs.
2024-01-22 20:37:53 -05:00
James Shubin
e946b39960 engine: graph: Set closed flag when resume signal closes
We had this backwards. Woops...
2024-01-22 20:37:53 -05:00
James Shubin
0711d05232 engine: resources: http: Improve groupability of resources
This improves groupability of http resources so that we can have fancier
kinds!
2024-01-22 20:37:53 -05:00
James Shubin
9dd5dfdde2 engine: graph: autogroup: Improve the autogrouping algorithm
This improves the autogrouping algorithm to support hierarchical
autogrouping. It's not guaranteed to work if we replace the reachability
grouper with something more efficient, but it's good enough for now.
2024-01-22 20:37:53 -05:00
James Shubin
024aa60209 lang: parser: Allow edges of resources with colons
Due to a small copy-pasta bug in the parser, resource edges with kinds
containing colons didn't work. This fixes the mistake and adds a test.
2024-01-22 19:40:15 -05:00
James Shubin
32916f9a6f engine: resources: Receive keys should match mcl, not golang
The capitalization of these keys was wrong and they weren't getting
seen. Add a test as well.
2024-01-22 19:40:15 -05:00
James Shubin
b670bb8d2c engine: resources: http: Recurse on Init and Watch
Make this resource handle sub-resources more powerfully.
2024-01-22 19:16:28 -05:00
James Shubin
1abf6547ff engine: resources: http: Improved CheckApply for grouped resources 2024-01-22 19:16:28 -05:00
James Shubin
3739fa401e engine: resources: Refactor HTTP constants 2024-01-22 17:43:24 -05:00
James Shubin
377d62999f engine: resources: The kv resource can set un-mapped values
Previously the resource could only set values in a per-hostname
namespace, but for single, user-managed values, we'd like to be able to
control things entirely. Now this resource can do that.
2024-01-22 17:43:24 -05:00
James Shubin
6f1f69683d lang: funcs: core: world: Add a function to get world values
Useful for following single values that we set.
2024-01-22 17:43:24 -05:00
James Shubin
b4bb4ca397 engine: resources: Update kv error messages 2024-01-22 17:43:24 -05:00
James Shubin
6b4fb434da engine: resources: Rename var
More accurate.
2024-01-22 17:43:24 -05:00
James Shubin
9e70f53afa engine: resources: Add missing apply guard 2024-01-22 17:43:24 -05:00
James Shubin
81885dec63 engine: resources: Add some spacing 2024-01-22 17:43:24 -05:00
James Shubin
f59f3c3c83 lang: funcs: core: convert: Add format and parse for bool 2024-01-22 17:43:24 -05:00
James Shubin
18c66ae7ac lang: funcs: core: Remove unused struct fields and bool
Old copy-pasta misc!
2024-01-22 17:43:24 -05:00
James Shubin
d01c168450 lang: Add per-test config with count maximums
Some of our special tests can only be run once per `go test` invocation.
That is, using the test -count flag will cause a guaranteed failure
since we depend on a global being initialized only once as part of that
test.

This adds a per-test config option so that a user can specify to never
run a particular test more than once. This lets us continue to use the
-count flag with the test suite, without it causing some tests to fail.
2024-01-22 16:55:45 -05:00
James Shubin
f0a4a9e3c4 lang: Don't print test comments if empty 2024-01-22 16:53:47 -05:00
James Shubin
78e59a9400 github: New versions of github actions 2024-01-22 15:52:49 -05:00
James Shubin
a8f194259b legal: Happy 2024 everyone...
Done with:

ack '2023+' -l | xargs sed -i -e 's/2023+/2024+/g'

Checked manually with:

git add -p

Hello to future James from 2025, and Happy Hacking!
2024-01-22 15:52:49 -05:00
James Shubin
c39fdcb8ac lang: funcs: dage: Have the compiler verify expectations
This engine should fulfill the GraphAPI interface. Specify that here in
case it's not obvious.
2024-01-22 15:52:49 -05:00
James Shubin
26e46a6e18 lang: funcs: txn: Add a simple graph implementation
This fulfills the GraphAPI that we use.
2024-01-22 15:52:49 -05:00
James Shubin
4ad7edc35e lang: funcs: txn: Move transaction code to a new package
This also makes it public, although it is designed for internal use
only.
2024-01-22 15:38:18 -05:00
James Shubin
28eacdb2bb lang: funcs: ref: Move reference counting code to a new package
This also makes it public, although it is designed for internal use
only.
2024-01-22 15:38:18 -05:00
James Shubin
66b826a8e1 pgraph, lang: funcs: dage: Add an awkward graphviz interface
This refactors the code to make it not depend on the specific engine.
2024-01-22 15:35:21 -05:00
James Shubin
20c8a856a2 lang: funcs: dage: Add a Graph method to improve the API
This also adds it to the GraphAPI so that users of it can pull out a
graph when needed. It's not likely to be used by the dage engine
consumers.
2024-01-22 15:09:32 -05:00
James Shubin
dd20bd5486 lang: ast: Improve ordering to eliminate false positives
The Ordering and DAG detection code is challenging because we need
Ordering to do SetScope, but Ordering itself needs to know about scopes.
This improved variant should hopefully catch all the scenarios of
identically named variables causing invalid loops.

Co-authored-by: Samuel Gélineau <gelisam@gmail.com>
2024-01-22 13:19:10 -05:00
James Shubin
f8077d9fc4 lang: ast: Add a util function for node map copying 2024-01-22 13:07:17 -05:00
Samuel Gélineau
16dfb7b5f5 lang: Add more test cases related to future graph optimization
We are planning to implement an optimization in which some function
calls are compiled to a single static graph rather than to a CallFunc
which dynamically creates a sub-graph at runtime. These test cases
exercise corner cases for which it would be theoretically possible to
use a static graph, but which we might not be able to optimize.
2024-01-20 15:01:56 -05:00
James Shubin
aae0e16350 lang: ast, parser, interfaces: Implementation of nested class sugar
This implements a new type of syntactic sugar for the common pattern of
a base class which returns a child class, and so on. Instead of needing
to repeatedly indent the child classes, we can instead prefix them at
the definition site (where created with the class keyword) with the name
of the parent class, followed by a colon, to get the desired embedded
sugar.

For example, instead of writing:

class base() {
	class inner() {
		class deepest() {
		}
	}
}

You can instead write:

class base() {
}
class base:inner() {
}
class base:inner:deepest() {
}

Of course, you can only access any of the inner classes by first
including (with the include keyword) a parent class, and then
subsequently including the inner one.
2024-01-14 17:10:31 -05:00
James Shubin
741a71b490 lang: parser: Clean up the parser to avoid parsing res separately
This cleans up some cruft in the parser, and makes it easier for us to
extend this to future colon separated identifiers.
2024-01-14 17:10:31 -05:00
James Shubin
44ee578a3a lang: parser, ast, interfaces: Implement include as
This adds support for `include as <identifier>` type statements which in
addition to pulling in any defined resources, it also makes the contents
of the scope of the class available to the scope of the include
statement, but prefixed by the identifier specified.

This makes passing data between scopes much more powerful, and it also
allows classes to return useful classes for subsequent use.

This also improves the SetScope procedure and adds to the Ordering
stage. It's unclear if the current Ordering stage can handle all code,
or if there exist corner-cases which are valid code, but which would
produce a wrong or imprecise topological sort.

Some extraneous scoping bugs still exist, which expose certain variables
that we should not depend on in future code.

Co-authored-by: Samuel Gélineau <gelisam@gmail.com>
2024-01-14 17:08:51 -05:00
James Shubin
f92f34dc54 lang: The in keyword can't be a variable name
At least for now until we figure out some fancier parser magic. (If
at all possible!)
2024-01-12 18:22:00 -05:00
James Shubin
dff9c9ad22 lang: ast: Add more validation on struct fields
Catch more errors early, just in case.
2024-01-12 18:11:39 -05:00
James Shubin
acae7eccc4 lang: interfaces: Expand docs on compiler constants
Improve the docs to make things clearer.
2024-01-07 18:44:58 -05:00
James Shubin
837739d7dd pgraph: Implement FilterGraph in terms of FilterGraphWithFn
Simplify the code base by having one core implementation.
2024-01-07 18:44:58 -05:00
James Shubin
2567303fd7 pgraph: Add FilterGraphWithFn implementation
This lets us filter more programmatically.
2024-01-07 18:44:58 -05:00
James Shubin
4939ae1a2f pgraph: FilterGraph doesn't need a name arg
It's not being consumed anywhere, so remove it. If you really want to
rename the graph, this can be done as a second step.
2024-01-07 18:25:50 -05:00
James Shubin
58607e2ca0 pgraph: Filter graph should not include unfiltered vertices
This adds a test for another sneaky aspect of filter graph: It should
not pull in a vertex because another across the edge was included. If we
want this behaviour, then we need a different function or a config
option for this filter function.
2024-01-07 18:24:56 -05:00
James Shubin
cb021e3a25 pgraph: Filter graph should preserve vertices without an edge
There was a bug in filter graph that caused it to not preserve any
vertices that weren't bound to another with an edge. This fixes the bug
and adds a test too.
2024-01-07 17:56:02 -05:00
James Shubin
d61936c06d test: Add some simple commit body tests
I never remember the capitalization rules, so hopefully this helps me!
2024-01-05 16:57:42 -05:00
Samuel Gélineau
3553eb1f2a lang: ast: Fix data race in ExprSingleton
Init the mutex everywhere, but consider calling Init instead and
plumbing though the data input field in the future.
2024-01-05 16:27:41 -05:00
James Shubin
c4a9560d53 engine: local: Fix benign race
Use the mutex in a safer manner to eliminate the benign data race.
2024-01-05 15:50:43 -05:00
James Shubin
bc63b7608e engine: graph: Use an atomic bool instead of a mutex
The isStateOK variable can be accessed concurrently as these are
supposed to be "benign" races. As such, they need to be labelled as such
so that we don't hit some undefined compiler behaviour.

Here are five good references relating to "benign" data races in golang.

1) https://web.archive.org/web/20181022150257/https://software.intel.com/en-us/blogs/2013/01/06/benign-data-races-what-could-possibly-go-wrong

2) https://go.dev/ref/mem - "Informal Overview" section.

3) https://docs.oracle.com/cd/E19205-01/820-0619/gecqt/index.html

4) https://www.usenix.org/legacy/event/hotpar11/tech/final_files/Boehm.pdf

5) https://go.dev/doc/articles/race_detector

TL;DR: wrap your benign races with sync/atomic or eliminate them.
2024-01-05 15:50:42 -05:00
James Shubin
c2f508e261 engine: graph: Use an rwmutex around read/write of vertex state
This fixes two small races we had while simultaneously reading from and
writing to the vertex timestamp, and simultaneously writing to the
vertex isStateOK (dirty) flag.

They were actually "safe" races, in that it doesn't matter if the
read/write race got the old or new value, or that the double write
happened. The time sequencing was correct (I believe) in both cases, but
this triggers the race detector now that we have tests for it.
2024-01-02 18:18:34 -05:00
James Shubin
a07dc0a511 lang: interpolate, funcs: Add a concat function for faster interpolation
This adds a concat function which can be used directly by string
interpolation to avoid having to constantly unify the plus operator
which is much slower at this time.

The new monomorphisms changes caused type unification of a notable
example to go from ~25s to ~5m30s which was obviously not bearable. With
this fix, things are now down to ~6s.

This is an important optimization, but it's also a good reminder that
type unification of polymorphic functions needs to be improved in
general too.
2024-01-02 15:23:40 -05:00
James Shubin
d8db320722 lang: gapi: Prettier printing of type unification timing
This makes things a little nicer to visualize.
2024-01-02 13:26:51 -05:00
James Shubin
24054f905f lib, lang, docs: Add --only-unify option
This adds a new run flag for the lang frontend to exit immediately
following type unification. This makes it easier to use this as a step
in CI, and also to type the execution for performance comparison
reasons.
2024-01-02 13:06:19 -05:00
James Shubin
32ca815896 lib, docs: Small cleanups
Formatting and removing of old docs.
2024-01-02 13:04:49 -05:00
James Shubin
fa5949e191 lang: funcs: core: test: Make a new instance for each test
Since this special one_instance function uses global state, if it's
re-used in more than one test, this won't work since they still all use
the whole global state. Make new ones for each test.

This also breaks the count=2 feature (any number other than 1) when
running these, which is not ideal. Create a cleanup API that we can run
between tests to reset the global state.
2023-12-27 18:25:53 -05:00
Samuel Gélineau
1c0a98a0cc lang: ast: ExprBind is now monomorphic
This adds ExprTopLevel and ExprSingleton and ensures that ExprBind is
now monomorphic.

This corrects a previous design bug where it was not monomorphic and
would thus cause spawning of many more copies than necessary. In most
cases this was only harmful to memory and performance, and not
behaviour, since these functions were pure, and we didn't have a test
for this.

This also adds a bunch more tests. Most notably, the graph shape tests
generally produce smaller graphs now.

Lastly, a lambda cannot have two different types when used at two
different call sites. It is rare that this would be used, and when it
would make sense, there are easy workarounds to accomplish equivalent
goals.

This was mostly authored by Sam, James helped with some cleanup and
debugging.

Co-authored-by: James Shubin <james@shubin.ca>
2023-12-27 16:33:38 -05:00
James Shubin
9d208e8795 lang: Test more iter funcs without polymorphic id function
This tests the lambdas in more ways so that we are sure non-polymorphic
$id functions work the way we want.
2023-12-27 16:33:38 -05:00
James Shubin
72fe0cd6db lang: Polymorphic lambda functions aren't allowed for now
We don't allow a lambda variable to be polymorphic anymore. Of course
this isn't bad, but it makes things too difficult, at least for now.
It's also likely that we don't need this specific feature very often I
think.
2023-12-27 16:33:38 -05:00
James Shubin
734590b6bd lang: Add a test for duplicate functions called
We should not call either of these functions more than once for their
values. If we do, it means we have made a mistake with a compiler
optimization.

This is important, because otherwise if you had code like:

$x = random_password()

Then this would obviously be a problem. Thankfully, the situations where
functions generate unique data is rare, but it's probably something we
should take care of.
2023-12-27 16:33:38 -05:00
James Shubin
7cc231e8b9 lang: unification, interfaces: Add a skip invariant
This is a cleaner way of telling the type unifier that we don't want
this particular expression in the solution set at the end.
2023-12-27 16:33:38 -05:00
James Shubin
c2bf4ef7d4 lang: Test a top-level var explicitly
This makes sure the top-level scope is really seen.
2023-12-27 16:33:38 -05:00
James Shubin
439179e37f lang: Remove duplicate test
It's contents were identical to the stmtfunc-recursive-double.txtar
test.
2023-12-27 16:33:38 -05:00
James Shubin
c333cb542c lang: ast: Expr Param and Poly should not have values
Sometimes a recursive call through ExprVar's Value method would hit one
of these and return (nil, nil) which would throw off things.
2023-12-17 21:39:06 -05:00
James Shubin
6a6546db8d lang: ast: Save the non-polyfunc args
We probably want to save these args too in case.
2023-12-16 14:26:29 -05:00
James Shubin
a6d22a5a4b misc: Add editorconfig info for txtar tests 2023-12-12 00:24:41 -05:00
James Shubin
da5f94e62c test: Add a simple mcl formatting test
This should definitely get expanded overtime when we can have something
decent for formatting mcl code. For now, catch the easy things.
2023-12-12 00:13:27 -05:00
James Shubin
1d886f2995 test: Improve bash fmt test
This catches slightly more issues.
2023-12-12 00:13:27 -05:00
James Shubin
a6b6aa570e test: Fix small test warnings
These were issues we should be catching.
2023-12-12 00:01:29 -05:00
James Shubin
32b26a09c2 test: Don't error on unbound variable
We would get this error instead of a test error:

	test/util.sh: line 32: GITHUB_ACTION: unbound variable

This would happen when run locally.
2023-12-12 00:00:19 -05:00
James Shubin
8fcaa4abf2 lang: Output of tests should also show autogrouped resources
This makes our test cases slightly more complete.
2023-12-08 18:25:54 -05:00
James Shubin
18e1f08156 engine: graph: Allow send/recv to work with autogrouped resources
We've previously not received a value from within an autogrouped
resource. It turns out this would be quite useful, and so this patch
implements the additional plumbing and testing so that this works!

Testing that an autogrouped resource can still send values has not been
done at this time.
2023-12-08 18:18:17 -05:00
James Shubin
bf5cc63bc5 misc: Move to golang 1.20
Make sure to quote your 1.20 otherwise it shows up as 1.2 which is very
old!
2023-12-03 21:40:48 -05:00
James Shubin
271a94e0c7 engine: resources: Improve logging on resource errors
When a subcommand errors, it's helpful to see what it was.
2023-12-03 21:21:55 -05:00
James Shubin
c3f34db81e test: Remove new grep warning
With newer versions of grep, you now see:

	grep: warning: stray \ before "

This is because we had an erroneous backslash escaping a quote. Remove
it.
2023-12-03 18:24:20 -05:00
James Shubin
b7d8a769db lang: funcs: core: Add value functions
This adds a new series of "get*" functions which can read values from
the associated "value" resources. The key name of the function must
match the name value of the resource for things to work.

Type unification isn't yet perfect in these scenarios, so you should use
casually and with caution.
2023-12-03 18:24:20 -05:00
James Shubin
c05af3b9b3 lib, lang, engine: local: Add Value mechanisms to local API
This expands the Local API with the first (and in theory, only ever) API
for reading and writing simple values. This is a coordination point for
resources and functions to share things directly.
2023-12-03 18:24:20 -05:00
James Shubin
3e37a60635 lang: funcs: core: test: Fix small formatting issue 2023-12-03 18:24:20 -05:00
James Shubin
9d47b6843f engine, gapi, lang, lib: Plumb through new local API
This is a new API that is similar in spirit and plumbing to the World
API, but it intended for all local machine operations and will likely
only ever have one implementation.
2023-12-03 18:24:20 -05:00
James Shubin
12ffac1f06 lang: Add placeholders for future prefix passing 2023-12-03 18:02:00 -05:00
James Shubin
7991b4ab25 engine, lang, lib: Re-order for consistency
Some small cleanups to make room for future patches.
2023-12-03 18:02:00 -05:00
James Shubin
984fb702e5 engine: resources: Make test resource easier for debugging
Let some fields not get printed if we request it.
2023-12-03 18:02:00 -05:00
James Shubin
019b88cedf util: Add value to B64 serialization functions
These small utility functions are useful.
2023-12-03 16:52:01 -05:00
Samuel Gélineau
55932cf3c7 lang: ast: Clear env on ExprCall.Graph()
Fix the same bug in ExprCall.Graph().
2023-12-02 01:19:31 -05:00
Samuel Gélineau
1ff2c9bbd9 lang: ast: Clear env in ExprVar.Graph()
Fix the bug in ExprVar.Graph() described in the previous commit.
2023-12-02 01:19:31 -05:00
Samuel Gélineau
72235b0fd4 lang: ast: New test "clear-env-on-var"
This test detects a mistake which is easy to make: when making a
recursive call to the target of an ExprVar, it would be easy to
accidentally pass the environment, like we usually do with every other
recursive call. For variables, this is a mistake, because the lambda
parameters which are in scope where the variable is used must not be in
scope where the variable is defined.

In fact, ExprVar.Graph() currently makes this mistake. The test passes
anyway, because an earlier phase (SetScope) correctly clears the
environment and detects the problem before the Graph phase. Thus, this
test does not guarantee that all the phases correctly clear their
environment, it merely detects the unlikely case in which all the phases
make the same mistake.
2023-12-02 01:19:21 -05:00
James Shubin
b8d391024a engine: resources: Add a DoneCmd feature for exec resource
This makes a common pattern of `$cmd && echo done > /tmp/donefile` much
clearer and easier.
2023-11-28 15:24:22 -05:00
James Shubin
598bec0eab engine: resources: Clean up var naming and misc in exec 2023-11-28 15:24:11 -05:00
James Shubin
709bf7b246 engine: resources: Add missing WatchCmd UID
Not sure if this is even used anywhere, but might as well add it for
now.
2023-11-28 15:24:11 -05:00
James Shubin
8251c8f259 lang: funcs: core: Add a second panic signature
This modifies the panic feature to accept a boolean or a string. If true
or not empty, then it will cause the panic. This makes some of the error
code a little less ugly.
2023-11-28 14:43:39 -05:00
James Shubin
9c0bde0b29 lang: funcs, parser: Add improved panic magic
This is a newer implementation of the panic magic. I kept the old commit
in for posterity and to show the difference. The two versions are
identical to the end-user with one exception: the newer version doesn't
include a useless panic resource in the graph when there is no panic. In
this version, the panic function returns false and the if statement it's
the condition of, doesn't produce the resource within. On error, we
still consume the function in the if expression, and doing so causes
everything to shutdown.

The other benefit is that the implementation is much cleaner and doesn't
need the interpolate hack.
2023-11-28 14:40:21 -05:00
James Shubin
2cbce963b7 engine: resources, lang: funcs, parser: Add panic magic
It's valuable to check your runtime values and to shut down the entire
engine in case something doesn't match. This patch adds some magic
plumbing to support a "panic" mechanism.

A new "panic" statement gets transparently converted into a panic
function and panic resource. The former errors if the input is not
empty. The latter must be present to consume the value, but doesn't
actually do anything.
2023-11-28 13:49:31 -05:00
James Shubin
64e6e686e0 lang: funcs: core: math: Add minus1 function
This is a useful template hack until we get something better.
2023-11-28 11:28:49 -05:00
James Shubin
dad3458ece mod: Try and update things
No idea why things break. go.mod sucks.
2023-11-27 21:00:03 -05:00
James Shubin
e727e7adb6 lang: funcs: core: net: Add more mac fmt functions 2023-11-27 21:00:03 -05:00
James Shubin
af1c952700 lang: funcs: funcgen: Allow []string as inputs
This allows us to have the strings.Join function generated.
2023-11-27 21:00:03 -05:00
James Shubin
ce2f7112a3 lang: funcs: funcgen: Don't nest unnecessarily
Clean up the code.
2023-11-27 19:57:54 -05:00
James Shubin
4650ed01eb lang: funcs: funcgen: Refactor slightly for list input
This code had some design issues. We'll rewrite this eventually, but for
now, let's allow list inputs. Refactor this to make it possible.
2023-11-27 19:57:54 -05:00
James Shubin
0e2c73a36d engine: resources: Reduce logging for file resource 2023-11-27 19:57:54 -05:00
James Shubin
139fed40dd docs: Add a reminder about a common gotcha
When I hit it the second time, I decided it needed to get added here.
2023-11-27 19:57:54 -05:00
James Shubin
ee88161808 engine: graph, resources: Reduce and clean up logging
Make the output more usable.
2023-11-22 22:37:35 -05:00
James Shubin
04d54e8d82 engine: resources: Mask normal closing error
This could possibly be fixed upstream, but we'll workaround it for now.
2023-11-22 22:11:50 -05:00
James Shubin
c6c0d3d420 lang: funcs: dage: Move much logging behind debug flag
Most of this logging isn't useful for ordinary usage. Hide it for now.
Eventually when we have a fancy logging system (curses-like) we can
bring back more information on the state of everything.
2023-11-22 22:10:38 -05:00
James Shubin
f4d70068b1 engine: resources: Add a default NBP option to dhcp:server
This let's us do some overrides and is a good alternative since
send/recv+autogrouping doesn't work yet.
2023-11-22 20:22:50 -05:00
James Shubin
3fe5a8d0d6 mod: Update tftp libraries 2023-11-22 20:21:40 -05:00
James Shubin
29a124900c mod: Update dhcp libraries 2023-11-22 20:21:40 -05:00
James Shubin
6670407c9c engine: resources: Make net resource reversible
This allows the net resource to be reversible if we want it to be. It
only flips the network state from up to down or vice-versa.
2023-11-22 20:21:40 -05:00
James Shubin
641b6067cd engine: resources: Remove net config file on down
This cleans up the file we would have created when building the resource.
2023-11-22 20:21:40 -05:00
James Shubin
b1d61fa90b engine: resources: Ignore cleanup errors that don't matter
If the file is already gone, ignore it.
2023-11-22 20:21:40 -05:00
James Shubin
0c93cf2600 engine: resources: Add constants for state up and down
Safer and cleaner code.
2023-11-22 20:21:40 -05:00
James Shubin
53753c0932 engine: resources: Improve net error messages
This is a common case of parse failures for not using CIDR. Make it
cleaner for the user.
2023-11-22 20:21:40 -05:00
James Shubin
029cfaf1f8 engine, lib: Misc log and comment changes
Just misc polishing.
2023-11-22 20:21:40 -05:00
James Shubin
66edc39ccb engine: resources: Improve exec docs
So that I don't hit this too often.
2023-11-22 20:21:40 -05:00
James Shubin
3f42f785ac engine: resources: Handle recurse field more appropriately
This smoothes out some edge cases. We should add comprehensive tests in
the future.
2023-11-22 20:21:40 -05:00
James Shubin
fd4d32351b engine: resources: Refactor file mode code
Small cleanup in old code.
2023-11-22 20:21:40 -05:00
James Shubin
75d3895e84 engine: Improve logging for resources and exec
This makes things a bit clearer and fixes some silly formatting bugs.
2023-11-22 20:21:40 -05:00
James Shubin
d022d7f09e engine: resources: Rename the value field in value resource
Having a different name makes it more obvious, and also leaves us open
to add a string or bool field for more explicit typing.
2023-11-18 15:15:33 -05:00
James Shubin
88b414b9a3 engine: resources: Add new value resource
This is a strange resource which is probably most useful for passing
values between scopes. It supports a variant resource field, and should
only be used as a last resort and if you know exactly what you're doing.
2023-11-18 14:23:31 -05:00
James Shubin
47c441ba40 lang: Test the resource engine briefly
We run the resource engine once and look at its values. This is useful
for testing send/recv in particular.

The converger code is probably not working properly. We'll look into
that subsequently if this gets used a lot.
2023-11-18 13:45:46 -05:00
James Shubin
7105e38544 engine: util: Remove the ValueOf hack
We've got something in the core utility function now.
2023-11-18 13:45:46 -05:00
James Shubin
4b0cdf9123 lang: types, engine: graph: Support pointer interfaces
If we have rare, but special *interface{} values in resource structs, we
should be able to handle them normally. It's really not recommended that
you use these unless you know exactly why they are useful.
2023-11-18 13:35:52 -05:00
James Shubin
1c9fdc79c0 engine: util: Add a workaround for printing special resources
This let's our tests compare strings of interface fields!
2023-11-18 13:35:52 -05:00
James Shubin
90d04990ca engine, lang: Add an AST test that looks at fields too
This gives us more options for testing when we need those kinds of more
extensive resource examination features.
2023-11-18 13:35:52 -05:00
James Shubin
aa001ed2dc engine: graph: Improve error message for rare case
Not sure if we ever hit this, but we should fix it up since I noticed
it.
2023-11-18 13:35:52 -05:00
James Shubin
ce1c37dbca lang: Rename a var for consistency
This is going to get used by a future package.
2023-11-15 17:55:47 -05:00
James Shubin
3a3bc568b3 engine: Allow more send/recv cases
This allows the other part of this. This still needs to be improved, but
it unblocks some use-cases.
2023-11-12 17:50:07 -05:00
James Shubin
b048b2684b engine, lang: Allow resources with a field of type interface
This lets us add a resource that has an implementation with a field
whose type is determined at compile time. This let's us write more
flexible resources.

What's missing is additional type checking so that we guarantee that a
specific resource doesn't change types during run-time.
2023-11-12 17:25:03 -05:00
James Shubin
9a1a81925e engine: Improve error messages for ambiguity
This appears twice, make each one unique.
2023-11-12 17:09:50 -05:00
James Shubin
e38f3cc12c lang: types: Let the fancy TypeOf method match interfaces
This adds the functionality to match on interface kinds.
2023-11-12 16:14:38 -05:00
James Shubin
4fb356b19b engine: graph: Refactor the send/recv code to inline it
This is useful subsequently.
2023-11-12 16:13:04 -05:00
James Shubin
cb6b8a9670 engine: util: Refactor code for future changes
This makes things a bit cleaner and easier to follow and accepts future
patches easier.
2023-11-12 15:26:44 -05:00
James Shubin
efc5237265 lang: types: Improve error handling of value code
We can give a bit more feedback and try and also catch a rare scenario.
2023-11-12 15:24:16 -05:00
James Shubin
cb999af653 engine: Improve debugging for invalid send/recv res
If we don't implement send/recv properly, at least give the user more
information as to where.
2023-11-12 15:21:08 -05:00
James Shubin
8d7d2fb1f1 engine: resources: Add missing struct tags for password res
Just noticed these.
2023-11-04 16:59:09 -04:00
James Shubin
eaff060bf9 lang: ast: Add missing backtick
Purely a formatting fix.
2023-11-04 16:57:41 -04:00
James Shubin
6cc5adcd25 lang, test, examples: lang: Use new syntax for structlookup
Sugar makes the world go round.
2023-11-04 14:52:19 -04:00
James Shubin
233625db20 lang: parser: Lexer should allow in keyword as a variable name
We move it downwards to allow this case. Whether we want to allow this
long-term or not is to be decided.
2023-11-04 14:37:19 -04:00
James Shubin
96093984e4 test, examples: lang: Add new syntax for map lookup
Sugar at last!
2023-10-17 15:32:13 -04:00
James Shubin
ea0af4dc43 lang: parser: Add lookup functions to lexer/parser
This seems to make things work. I'm worried I might have an ordering bug
where we might choose the wrong precedence if we write ambiguous code
somehow, but at least for now, let's commit this and move on. Hopefully
the nonassoc stuff is actually correct.
2023-10-17 15:21:18 -04:00
James Shubin
8fa5241a13 lang: funcs: Add struct lookup with an optional field
This adds an interesting version of the struct lookup function. In the
situation where we can't type-check the field name, it will use the
optional value passed in. This makes it easy to write a function that
will pull in the desired value, even as the input struct changes type
between compilations, without having to re-write your code.

It's structurally different from the other default lookup functions,
which is why it is named differently.
2023-10-17 15:21:18 -04:00
James Shubin
e38eb43955 lang: funcs: Add core lookup functions
These versions don't take defaults and instead return the zero value if
there is an issue.
2023-10-17 15:21:18 -04:00
James Shubin
3b46e88734 lang: types: Improve docs for New and initialize the list
If we use this to generate a zero value, we want to make sure it's
completely initialized in case we use it subsequently. We also improve
the docs at the same time.
2023-10-17 15:21:18 -04:00
James Shubin
d6a58f33f3 lang: funcs: Add lookup default func for list or map
This version requires you specify the default value.
2023-10-17 15:21:18 -04:00
James Shubin
d1c15bd0b7 lang: funcs: Rename the lookup functions
This will make things more consistent with future use.
2023-10-17 15:21:18 -04:00
James Shubin
1dc6ebbffc lang: funcs: Fix one small typo and one small logic bug
I don't think the logic bug had any negative effect, but let's be safe.
2023-10-17 15:21:18 -04:00
James Shubin
ee1e07f3d7 lang: funcs: Rename T2 to T3 for future consistency
This makes things more symmetrical with the maplookup function.
2023-10-17 15:21:18 -04:00
James Shubin
9e7b7fbb3a lang: parser: Define variable identifier in the parser
Instead of in the lexer. I think this simplifies things and gives the
parser more information which should hopefully make it easier to parse
without shift/reduce conflicts.
2023-10-17 15:21:18 -04:00
James Shubin
04fd330733 test: Improve parser indentation test
We need to be fancier with our regexp now that pipes are in the grammar.
Don't match them in comments of course!
2023-10-17 15:21:18 -04:00
James Shubin
361b671799 lang: Clean up test 2023-10-11 17:56:25 -04:00
James Shubin
b8f8ec8aa3 lang: parser, funcs: Take out the built-in history function
It's a normally named function for now, until we think of a common
package to move it into.

Hopefully this makes improving the lexer and parser easier for now. We
can consider bringing it back if needed.
2023-10-09 17:19:08 -04:00
James Shubin
05d570d250 lang: parser, funcs: Change the logical operators to OR, AND, NOT
This makes it easier to read for some, and easier to parse for us. This
also frees up more characters to use elsewhere.
2023-10-09 17:19:08 -04:00
James Shubin
e8f11286dc lang: parser: Sort things in the standardized order
Nitpick fix.
2023-10-09 17:16:06 -04:00
James Shubin
48bc5637a0 misc: Add the mcl type to ackrc for ack users
Now you can `ack --type=mcl` to find .mcl files!
2023-10-09 16:43:50 -04:00
Samuel Gélineau
7bfbe264c7 authors: Adding myself to the authors file 2023-10-07 12:50:19 -04:00
James Shubin
e37876afec engine: resources: Small formatting cleanups 2023-10-07 11:44:47 -04:00
James Shubin
02153356de lang: funcs: core: net: Tidy up test 2023-10-03 18:45:17 -04:00
James Shubin
a686c5f9aa lang: parser: Add new tests for deeper modules
You can have a module that's nested two deep or more.
2023-10-03 16:36:37 -04:00
James Shubin
ab7eb0a293 util: password: Reverse the nonblocking stdin
It's not clear if this is absolutely necessary or not, but it probably
doesn't hurt. It's not clear if there is a way to read the previous
state before running this. Of course this isn't really thread-safe, so
use at your own risk.
2023-09-29 22:13:14 -04:00
James Shubin
738485a655 etcd: Add world API changes for watching member status
This adds some useful functionality so that anyone with access to the
world API, can learn information about the changing etcd cluster it's
using.
2023-09-29 20:46:36 -04:00
James Shubin
0c751ea14f etcd: util: Move etcd utils into separate package 2023-09-29 20:46:36 -04:00
James Shubin
01f249d484 util: password: Add a cancellable ReadPassword style functions
It was non-trivial to do this, so I put it into a library. Strangely I
couldn't directly wrap the ReadPassword function from the
golang.org/x/term package, as it wouldn't unblock for some reason.
2023-09-29 20:46:36 -04:00
James Shubin
53f9f35233 docs: Misc fixes 2023-09-25 18:53:54 -04:00
James Shubin
9a890c7625 misc: More script notes added 2023-09-25 18:53:51 -04:00
James Shubin
2bc23c468e lang: funcs: core: iter: Finish map function
This was the goal all along. Proper iteration without for loops.

Yay!

Co-authored-by: Samuel Gélineau <gelisam@gmail.com>
2023-09-25 18:53:31 -04:00
James Shubin
6e1cde815c lang: ast: Implement lambdas
This is a big giant patch that implements the AST part of lambdas!

I don't know how Sam is able to understand the AST so well, but he does,
and we're all grateful for it. Most of this code was written by him.

Co-authored-by: Samuel Gélineau <gelisam@gmail.com>
2023-09-25 18:52:26 -04:00
James Shubin
d1f1ed8957 lang: funcs: simple, simplepoly, operator: Misc fix ups for lambdas
These core functions need some small fixups to bring us up to our
lambdas branch.

Co-authored-by: Samuel Gélineau <gelisam@gmail.com>
2023-09-25 18:52:05 -04:00
James Shubin
a1a23b66c8 lang: funcs: structs: Core function struct modifications
These changes are needed for the lambdas implementation.

Co-authored-by: Samuel Gélineau <gelisam@gmail.com>
2023-09-25 18:51:51 -04:00
James Shubin
b4a8d0d783 lang: interfaces, ast: Add debug methods for graphing scope
This can occasionally help in our debugging of the scope graph.

Co-authored-by: Samuel Gélineau <gelisam@gmail.com>
2023-09-25 18:51:33 -04:00
James Shubin
fcc76809e3 lang: interfaces: API changes for lambdas
Co-authored-by: Samuel Gélineau <gelisam@gmail.com>
2023-09-25 18:51:25 -04:00
James Shubin
1eae5cf272 lang: interfaces: Remove indexes
These indexes might be useful later, but they are currently not used.
2023-09-25 18:51:22 -04:00
James Shubin
ad1ee0f3dc lang: ast: Remove unused scope test
Maybe this will come back for function values, but it's not needed for
now.
2023-09-25 18:51:13 -04:00
James Shubin
0f99bfe7c4 lang: types: Add full func value
This needs to be in a separate package for now due to import cycles.

Co-authored-by: Samuel Gélineau <gelisam@gmail.com>
2023-09-25 18:51:07 -04:00
James Shubin
dbba49540e lang: types: Export the base value for others
This lets others implement this more easily.
2023-09-25 18:51:03 -04:00
James Shubin
5440ef7eb2 lang: types: Misc function changes
This will probably change up again in the future, but this is what we
need for now to make lambdas work.

Co-authored-by: Samuel Gélineau <gelisam@gmail.com>
2023-09-25 18:50:53 -04:00
James Shubin
8d63b98212 lang: Add a bunch of new language tests
These test both graph shape consistency and single value outputs.
Eventually we want to make the graph shape tests more precise, and also
verify specific outputs how it used to be. For now, this is okay.

Co-authored-by: Samuel Gélineau <gelisam@gmail.com>
2023-09-25 18:50:43 -04:00
James Shubin
d4b1e8f1be lang: Core language and GAPI changes
These changes help plumb things in more easily for the lambdas work.
2023-09-25 18:50:07 -04:00
James Shubin
47d2a661bc lang: interfaces, funcs: Add a new graph engine called dage
This adds a new implementation of the function engine that runs the DAG
function graph. This version is notable in that it can run a graph that
changes shape over time. To make changes to the same of the graph, you
must use the new transaction (Txn) system. This system implements a
simple garbage collector (GC) for scheduled removal of nodes that the
transaction system "reverses" out of the graph.

Special thanks to Samuel Gélineau <gelisam@gmail.com> for his help
hacking on and debugging so much of this concurrency work with me.
2023-09-25 18:48:40 -04:00
James Shubin
7f46bcef61 lang: types: Loop in deterministic order
This makes some traces and reproducing of bugs slightly more
deterministic.
2023-09-25 18:48:38 -04:00
James Shubin
4fd90b5d52 lang: types: Avoid a panic if it's not settable
We should check this for safety. An error is better than a panic. If we
try to set an unexported field, this would panic. We should prevent
being able to even type unify that though!
2023-09-25 18:48:33 -04:00
James Shubin
28c206da18 lang: ast: structs: Remove value lookup from if expr type
This is really not needed and makes our future API changes to Value more
difficult.
2023-09-25 18:48:28 -04:00
James Shubin
5e58735bb3 lang: funcs: Add listlookup function
This looks up a value in a list from an integer index.
2023-09-25 18:48:21 -04:00
James Shubin
c06c391461 lang: Update the Build signature to return a type
This returns the type with the arg names we'll actually use. This is
helpful so we can pass values to the right places. We have named edges
so you can actually see what's going on.

Co-authored-by: Samuel Gélineau <gelisam@gmail.com>
2023-09-25 18:48:15 -04:00
James Shubin
31c7144fff lang: ast, interfaces, interpret: Change the Output sig
This plumbs through the new Output method signature that accepts a table
of function pointers to values and relies on the previous storing of the
function pointers to be used for the lookup right now. This has the
elegant side-effect that Output generation could run in parallel with
the graph engine, as the engine only needs to pause to take a snapshot
of the current values tables.

Co-authored-by: Samuel Gélineau <gelisam@gmail.com>
2023-09-25 18:48:07 -04:00
James Shubin
9175d26b3b lang: ast, interfaces: Plumb in the Graph sig changes and value pointers
The Graph signature changes are needed for future function work, and it
also fits in nicely with the need for storing the value pointer for each
function node. These are used to later extract values during the Output
stage.

Sam deserves all of the credit for realizing both of these points and
convincing me to make the change! It worked out great, cheers!

Co-authored-by: Samuel Gélineau <gelisam@gmail.com>
2023-09-25 18:47:58 -04:00
James Shubin
f1d6f70cd2 lang: interfaces: Partially change Graph signature
This makes some of the Graph sig changes to prepare the code for proper
functions. The remaining bits will happen later.

Co-authored-by: Samuel Gélineau <gelisam@gmail.com>
2023-09-25 18:46:30 -04:00
James Shubin
b27b809ea9 lang: funcs: Remove old engine code
This gets rid of the old engine implementation which can only run a
static graph. We'll need a changing graph for lambdas.
2023-09-19 14:45:58 -04:00
James Shubin
1ec837089e lang: interpolate: Rename interpolate functions to please linters 2023-09-19 14:45:44 -04:00
James Shubin
089267837d engine: graph: Add a lock around metas access
We forgot about the concurrent writes. This should fix that.
2023-09-19 13:46:20 -04:00
James Shubin
c851322826 engine: resources: Fix up test resource
Woops...
2023-09-19 09:12:59 -04:00
James Shubin
1b12db92ab util: socketset: Add missing return
If our machine has that pipe busy, don't panic the test. (We do want it
to fail though!)

We're also more careful to check nil object just as a convenience to
help programmers.
2023-09-19 06:50:26 -04:00
James Shubin
6c0775ba59 engine: resources: Consider passing in funcs if possible
Let this sit in the test resource for now.
2023-09-04 16:33:05 -04:00
James Shubin
0d381e4c91 engine: graph: Handle the back poke differently
A back poke is the deferral or delay of a Process/CheckApply. This is
because we notice that we're not truly ready to CheckApply due to some
timestamp issue. When Process errors, we should accept that, but not
treat it as a success.
2023-09-04 15:22:14 -04:00
James Shubin
7ccda7e99b engine: Add a ctx to the CheckApply API
This is just a rough port, there are lots of optimizations to be done
and lots of timeout values that should be replaced by a new timeout meta
param!
2023-09-02 01:34:42 -04:00
James Shubin
567de2e115 engine: Use a struct instead of a compound string for resUID
This struct combination of resource kind+name is suitable for use as a
unique map key.
2023-09-02 01:34:42 -04:00
James Shubin
cf49a9a6e7 engine: graph: Give retry channel its own signal
This just makes the copy+pasting less confusing.
2023-09-02 00:59:48 -04:00
James Shubin
41493993a2 engine: resources: Respect the apply variable
Woops!
2023-09-01 22:51:44 -04:00
James Shubin
1d10f85c28 engine: graph: Move misplaced comment 2023-09-01 22:13:34 -04:00
James Shubin
d3d84524f5 engine: graph: Improve limit output
Just show the milliseconds, and round it slightly.
2023-09-01 21:57:05 -04:00
James Shubin
cc04221516 engine: graph: Allow pause/resume while in retry or limit
The retry and limit "satellite" event loops didn't allow pausing or
resuming, and instead you needed to wait until either was done before
you could pause.

The downside of this patch is that for very fast graph transitions, we
wouldn't be really obeying the limits anymore, however now that we have
per resource kind+name uid, we can persist the limits across graph swaps
if we want to.

Most importantly, this allows us to exit entirely when we're stuck in
one of these satellite loops.
2023-09-01 21:57:05 -04:00
James Shubin
f9bc50e262 engine: Retry should be stateful and add RetryReset
Make the retry meta param a bit more sane now that we can persist it
between graph switches. This also unblocks us from pausing during retry
loops.
2023-09-01 21:57:05 -04:00
James Shubin
9545e409d4 engine: Create a resource kind+name specific stateful store
This adds a meta state store that is preserved between graph switches if
the kind and name match. This is useful so that rapid graph changes
don't necessarily reset their retry count if they've only changed one
resource field.
2023-09-01 21:57:05 -04:00
James Shubin
07bd8afc4a engine: graph, metaparams: Default rewatch to false
The frequency of graph changes makes it unlikely that you want this
enabled on most resources by default.
2023-09-01 19:02:11 -04:00
James Shubin
9ac8d7ec49 test: shell: Disable flaky tests
I think this is related to timing and slow CI, but I really don't want
to waste time with old shell yaml tests.
2023-09-01 17:17:59 -04:00
James Shubin
b4bdc8adee github: Don't fail other jobs just because one failed
We like fast CI, but we prefer to have more information about the
failures.
2023-09-01 17:08:59 -04:00
James Shubin
8299c04fc6 engine: graph, util: Clean up error printing
We should improve on this more, but at least as a quick fix, stop
splitting the error across two lines. This makes the logs really ugly.
2023-09-01 16:53:40 -04:00
James Shubin
0b1b0a3f80 engine: graph: Don't deadlock on error
This simplifies the pause mechanism and also avoids a deadlock on error.
If the Worker shuts down completely, but before we've been removed from
the graph, then an attempted pause would deadlock if we didn't have an
escape hatch here.

This removes the unnecessary ack mechanism now that we have a
synchronous channel send to represent the pausing, rather than an
asynchronous channel closing.
2023-09-01 16:53:40 -04:00
James Shubin
2773a621a2 engine: graph: Cleanup pause/resume code
There's always the fear that there is either a panic or a deadlock in
the highly concurrent engine resource code. I have not seen one recently
and I've been running some pretty concurrent tests. In the meantime, and
with my hopefully improved knowledge of concurrency, I decided to
rewrite some of the "uglier" parts of the engine. I think it is a lot
clearer now, and much less likely that there is a concurrency issue.

This has been tested by running the examples/lang/fastcount.mcl example.
2023-08-30 21:58:38 -04:00
James Shubin
2edae22a65 lang: funcs: core: test: Add a new test functions package
This adds a new test functions package and also a new "fastcount"
function which counts up from zero as fast as is possible. You probably
don't want to use this in production, but it is useful for performance
and deadlock testing the resource and function engines.
2023-08-30 21:58:38 -04:00
James Shubin
c06cf44fd7 lib, engine: graph: Rename the Close method 2023-08-30 21:38:01 -04:00
James Shubin
7288e5d4a4 engine: graph: Improve documentation on concurrent use
Certain other methods should not be called concurrently, but this only
documents the most important cases.
2023-08-30 21:38:01 -04:00
James Shubin
b62f501745 engine: graph: Fix typo
We actually now cleanup instead of closing. It's semantically slightly
different, so be consistent with the error message.
2023-08-30 21:38:01 -04:00
James Shubin
c199a51eeb readme: Travis is not really being used anymore 2023-08-30 13:01:59 -04:00
James Shubin
b60d222c81 integration: Port tests to use standalone etcd
This ports the integration tests to use a standalone etcd server instead
of depending on the flaky elastic etcd clustering. Hopefully we will
polish and/or reimplement that at some point in the future, but at least
for now let's make things reliable.
2023-08-30 03:44:19 -04:00
James Shubin
fb57fb714a main, lib: Build in standalone etcd
Standalone etcd is useful for when we don't want to use the embedded
version to make it easier to deploy somewhere or for testing.

This pulls in about the same amount of code since we already embedded
etcd previously. Since the embedded etcd feature of mgmt is not very
stable, we'll add this for now.
2023-08-30 03:44:12 -04:00
James Shubin
bc390088b3 etcd: Don't unset our only endpoint
When mgmt is in etcd-client-only mode and using an external etcd server,
we don't want to unset our only known endpoint since this would deadlock
our etcd client since it can't connect to anyone. This could have
happened because a plain etcd server didn't set any endpoints to follow,
and as a result we noticed it was empty and decided to use that instead.

To workaround this issue on an earlier version of mgmt, you would have
had to run:

etcdctl put /_mgmt/endpoints/etcd http://localhost:2379

to set this magic key on the initial etcd server.
2023-08-30 03:05:22 -04:00
James Shubin
532e2ec8e1 etcd: Chooser shouldn't be used with --no-server
We forgot to check this in Init().
2023-08-30 03:05:22 -04:00
James Shubin
6f4b6cec7e make: Add a flag to disable trimpath except on release
This is kind of magic now. Use .envrc in dev environments to disable it,
except during releases of course!
2023-08-30 03:05:22 -04:00
James Shubin
0b972c5b4b lang: funcs: core: fmt: Add more invariants to printf
These don't seem to change the number of passing tests, but add them in
case we hit some situation that benefits from them.
2023-08-29 19:33:46 -04:00
James Shubin
45e126bf07 lang: funcs: core: fmt: Allow dynamic format strings
There are many reasonable cases where we might want to allow a dynamic
format string. Support that situation by adding the new invariants that
are needed for those cases.
2023-08-29 19:33:45 -04:00
James Shubin
318f28affd lang: unification, interfaces: Don't pass over generators
We were skipping over being fully consistent with all of the generator
invariants when running the solver. This allowed us to miss some of the
conditions that a generator might impose. Usually this caused us to be
"solved" when in fact we had an invalid program.
2023-08-29 19:33:45 -04:00
James Shubin
2214954c51 lang: unification: Error instead of panic
This is still a programming error, but not as dangerous for the edge
cases we might still have present, particularly with printf.
2023-08-29 19:33:45 -04:00
James Shubin
ec515f4fb5 make: Make builds more likely to be reproducible
As per https://go.dev/blog/rebuild if we include -trimpath then we'll
not store full build paths in the build artifacts.

We still have some CGO dependencies, but we'll look into those
separately.
2023-08-28 17:44:40 -04:00
James Shubin
3fb75707e7 lang: types: We should use the platform-dependent int instead
This could be a 32 or 64 bit sized int depending on arch.
2023-08-28 15:54:36 -04:00
James Shubin
c1850e0e20 engine: resources, lang: Set resource fields more accurately
There were some bugs about setting resource fields that were structs
with various fields. This makes things more strict and correct. Now we
check for duplicate field names earlier (duplicates due to identical
aliases) and we also don't try and set private fields, or incorrectly
set partial structs.

Most interestingly, this also cleans up all of the resources and ensures
that each one has nicer docs and a clear struct tag for fields that we
want to use in mcl. These are mandatory now, and if you're missing the
tag, then we will ignore the field.
2023-08-26 01:11:50 -04:00
James Shubin
b8d87e2d5a lang: funcs: core: fmt: Fix printf unification bug
We also add a backup fix to avoid a panic in case we ever hit a new
unification bug that lets something through, we can at least turn it
into a runtime issue. This adds a test as well.
2023-08-22 16:35:08 -04:00
James Shubin
1f53fd85b4 lang: unification: Add new eq to main list
We want to add the equality to the main list of equalities in case it is
needed somewhere else. This doesn't have any effect on any of our test
cases, but it doesn't seem to be harmful, and it could conceivably be
useful in the future. This is a separate commit in case we want to check
if this behaviour still holds true well into the future.
2023-08-22 15:56:04 -04:00
James Shubin
564c93ee21 lang: unification: Rename short vars
These two vars are used pretty deeply through this code, so rename them
to prevent any confusion.

We also switch from a range loop to a counter so that the loop list can
be changed while it's looping.
2023-08-22 15:55:55 -04:00
James Shubin
6769786241 lang: unification: Add equiv matching and resultant equalities
The set of initial invariants that we see might include:

	?8 = func(inputs []int, function  ?4) ?3
	?8 = func(inputs []int, function ?10) ?9
	?8 = func arg0   []int, arg1      ?6) ?7

From this we can infer that since they are all equal, that we also know
that ?4, ?10 and ?6 must also be equal. The same is true of ?3, ?9 and
?10. Those new equalities are sometimes necessary in order to complete
the full unification.

The second interesting aspect is when we have dissimilar equalities:

	?2  = func(x  ?1) str
	?4  = func(a int) ?5
	?10 = func(a int) ?11

In this example we also have an additional equality:

	?6 = ?2

From this and the above we can determine that ?2, ?4, ?6 and ?10 are all
equal. We only know about ?4, ?6, and ?10 from the direct relationship,
and we add in ?2 from the indirect (graph) relationship. These
relationships let us determine new information that ?5 and ?11 are both
str and that ?1 is an int.

Two important reminders:

1) Arg names don't have to match. It would impossible to build such a
   system where this was both possible, but also let us name our
   functions sanely.

2) None of this guarantees we won't find an inconsistency in our
   solution. If this is found, it simply means that someone wrote code
   which does not type check.
2023-08-22 15:55:48 -04:00
James Shubin
0946e860f1 lang: unification: Catch possible rare bug
When adding some likely incorrect invariants (while testing code) we hit
this panic. While it should not occur, it's probably better to catch it.
2023-08-22 15:55:44 -04:00
James Shubin
33366469a3 lang: unification: Add partial func knowledge to solved
This performs the same task as the previous two commits, but it also
adds partial information when solving a func.
2023-08-20 21:00:02 -04:00
James Shubin
7d29efeb33 lang: unification: Add more partial knowledge to solved
This performs the same task as the previous commit, but it also adds
partial information when solving a list, map, or struct.
2023-08-20 21:00:02 -04:00
James Shubin
f0e9af1cf5 lang: unification: Add partial knowledge to solved
When we were running our partial unification to compare similar looking
function invariants, we would sometimes learn new information, even if
it didn't entirely solve that invariant. When this new information was
learned, it's important that we globally mark it as solved so that
others can use it to complete their invariants. When it was a critical
piece of information, this would result in new information, which would
eventually lead to that original partial invariant that we learned that
information from to becoming solved.
2023-08-20 21:00:02 -04:00
Samuel Gélineau
aec8e1db2d lang: unification: Add type inference state debugging
This presents things in a more formal way for those who are more
familiar with standard type unification syntax.

Patch created by Sam, and cleaned up by James.
2023-08-20 21:00:02 -04:00
James Shubin
170fb64bfc pgraph: graphviz: Update our graphviz library
This makes things a bit easier to use. Especially when building fancy
graphs.
2023-08-08 22:36:52 -04:00
James Shubin
b134c4b778 lang: interfaces, funcs: Port Func API to new Stream signature
This removes the `Close() error` and replaces it with a more modern
Stream API that takes a context. This removes boilerplate and makes
integration with concurrent code easier. The only downside is that there
isn't an explicit cleanup step, but only one function was even using
that and it was possible to switch it to a defer in Stream.

This also renames the functions from polyfunc to just func which we
determine by API not naming.
2023-08-08 21:33:06 -04:00
James Shubin
6a06f7b2ea pgraph: Make vertex sorting more deterministic
If two vertices have the same String output, then sort by pointer string
instead. This makes display of certain graphs more deterministic.
2023-08-08 01:12:06 -04:00
James Shubin
5e0922395c util: Add a new sync primitive named brs
This adds the BoundedReadSemaphore mutex that I invented. It's not that
is necessarily particularly revolutionary, but it is useful. I wish I
had a better name for it, but I couldn't think of anything. It's fairly
obvious what it does, so if you have a suggestion of how to name it,
please do so!
2023-08-08 01:12:06 -04:00
James Shubin
514927c0b3 engine: resources, graph: Change the Close method to Cleanup
This also changes a few similarly named methods. Clearer what it's doing
in terms of cleanup vs. causing some action.
2023-08-08 01:11:29 -04:00
James Shubin
63f05e12ca engine: resources: net: Clean up weird error handling
Not sure how this happened.
2023-08-08 01:11:29 -04:00
James Shubin
b87d70c71c engine: resources: Move wait out of Close method
Watch should wait until everything is done before exiting.
2023-08-08 01:11:29 -04:00
James Shubin
3f403d34a4 etcd: world: Make sure to check if wg is nil or not
We never hit this as far as I know, but might as well be safe.
2023-08-08 01:11:29 -04:00
James Shubin
5d84e33be7 etcd: world: Add missing defer
Not sure how this snuck in!
I've now ran a quick grep across the code base, and I can't find any
similar mistakes.

ack '.Done()' | grep -v defer | grep -iv ctx # then check these
2023-08-08 01:11:29 -04:00
James Shubin
00a37b6289 engine: resources: config_etcd: Fix missing return
Woops!
2023-08-08 01:11:29 -04:00
James Shubin
963393e3d9 engine: graph, resources: Change Watch to use ctx
This is a general port. There are many optimizations and cleanups we can
do now that we have a proper context passed in. That's for a future
patch.
2023-08-08 01:11:29 -04:00
James Shubin
53a878bf61 engine: resources, graph: Change the done channel into a ctx
This is part one of porting Watch to context.
2023-08-08 01:11:29 -04:00
James Shubin
5eac48094b engine: resources: Disable broken docker test
It's not my fault it broke, afaict!
2023-08-08 01:11:29 -04:00
James Shubin
90e8e02b0c misc: Move to new golang version 2023-08-07 23:52:37 -04:00
Laurent Indermuehle
0dcfe027b0 make: Add a Dockerfile to build mgmt for Fedora 2023-07-11 14:26:02 +02:00
James Shubin
7d73c7fca9 lang: funcs: core: Clean up functions with constants 2023-06-29 19:30:23 -04:00
James Shubin
d37862b123 lang: funcs: Use more constants to build types 2023-06-29 15:31:12 -04:00
James Shubin
df9b338279 util: Handle negative values for NumToAlpha
This shouldn't be used often, but it makes writing some kinds of code
easier.
2023-06-29 15:15:37 -04:00
James Shubin
97c7d176f0 lang: funcs: Use constants for arg names 2023-06-29 14:59:05 -04:00
James Shubin
f966b1ae6a lang: funcs: simplepoly: Fix copy pasta error
Print the actual result.
2023-06-27 18:49:00 -04:00
James Shubin
8bc08d7716 lang: unification: Add a test for different func arg names
It's important that during unification that function types can unify
even if their arg names are different. (Their actual unnamed types must
be identical of course.) This adds a test to make sure we preserve this
behaviour.
2023-06-27 16:37:14 -04:00
James Shubin
ef4c0f961d lang: unification: Improve simple solver
This adds a new interpretation of the EqualityWrapFuncInvariant that can
combine two into one if their Expr1 fields are the same. We might know
different partial aspects of multiple functions. This also includes a
test. The test does not pass before this commit which adds it.
2023-06-27 16:37:14 -04:00
James Shubin
42840827f8 lang: unification: Add bare solver tests
This adds some simple test infra to test the solver directly without an
AST.
2023-06-27 16:37:14 -04:00
James Shubin
13b6b9de60 lang: types: Print the full function args
This makes it easier to see logs, but more importantly copy (well the
copy hack) doesn't erase arg names anymore.
2023-06-25 12:25:44 -04:00
James Shubin
74a21bab1a test: Update rule configuration
We got a warning about this.
2023-06-25 02:11:01 -04:00
James Shubin
d1c7770949 lang: parser, funcs, core, iter: Rename xmap to map
The map function previously existed as "xmap" because this was a
reserved word in the parser. This now adds a special case for this
identifier name.
2023-06-01 18:08:37 -04:00
James Shubin
66edf22ea3 lang: Port TestAstFunc1 to txtar format 2023-06-01 16:56:44 -04:00
James Shubin
8fffd10280 lang: Port TestAstFunc2 to txtar format
This ports TestAstFunc2 from our home-grown content storage system to
the txtar package. Since a single file can be used to represent the
entire folder hierarchy, this makes it much easier to see and edit
tests.
2023-06-01 16:56:44 -04:00
James Shubin
446dbde836 util: Add new blocked timer utility
This adds a new *BlockedTimer struct which can be used to run a function
or a printf after a short duration. It can be cancelled early. A short
blog post about the topic is available here:

https://purpleidea.com/blog/2023/05/26/blocked-select-logging-in-golang/
2023-05-26 01:19:13 -04:00
James Shubin
ef84a5a76a test: Rename var
Copy pasted from old script.
2023-05-12 12:10:46 -04:00
Kaushal
25a63956f3 test: Add golangci-lint test 2023-05-12 12:07:52 -04:00
Ofek Atar
eddd16283d engine: util: Add test for StructKindToFieldNameTypeMap function 2023-05-12 14:45:42 +03:00
James Shubin
c5efe7a17b lang, engine: Remove unneeded error wrapping
These situations basically never fail, and if they do, we certainly
don't need more context. This simplifies things a bit.
2023-04-20 18:02:40 -04:00
James Shubin
7075b8b973 pgraph: Add LookupEdge function
The new LookupEdge function lets you find which vertices are associated
with a particular edge, if any.
2023-04-20 15:45:46 -04:00
James Shubin
3f5957d30e docs: readme: Add new talk and clarify README
This should make it more obvious how to reach this content as it might
not have been particularly clear to new users where to find this
information.
2023-04-07 11:51:43 -04:00
James Shubin
bc29957d1e lang: Add a useful debug message to tests
When running this test a lot in series, it helps to have more obvious
debug output.
2023-03-31 19:29:52 -04:00
James Shubin
289835039a lang: Remove SetValue from the engine
This removes the calling of SetValue from the engine, and instead
replaces it with the Table() API. The downside is that this is likely
slower, and the current API with locking being exposed publicly is kind
of ugly. The upside is that this might make building the new engine
easier.

Future versions might remove locking from the API if we can avoid making
any accesses to expressions. Currently this happens within Logf/SafeLogf
which is our main (only?) usage at the moment. Logging could become
smarter perhaps. Alternatively, we might pass in a "setter" function
that gets called safely from within the engine. This could wrap SetValue
and the locking functions wouldn't be part of the public API.
2023-03-31 19:29:52 -04:00
James Shubin
b1e08ef231 util: deadline: Add a simple test for deadline and testing
This is a simple test/example placeholder if we want to remember how to
run deadlines with contexts for testing.
2023-03-24 16:53:42 -04:00
James Shubin
8a463767bf lang: funcs: Make printable function names more unique
This helps a lot with debugging.
2023-03-24 16:50:01 -04:00
James Shubin
c598e4d289 engine, etcd: Update code for latest gofmt fixes
Latest version of golang broken gofmt again...
2023-03-14 16:43:08 -04:00
James Shubin
a7624a2bf9 legal: Happy 2023 everyone...
Done with:

ack '2022+' -l | xargs sed -i -e 's/2022+/2023+/g'

Checked manually with:

git add -p

Hello to future James from 2024, and Happy Hacking!
2023-03-05 18:31:52 -05:00
James Shubin
d20fcbd845 lang: Fixup tests that broke
I should not have changed the error string when I unified the pgraph
library.
2023-03-03 14:47:30 -05:00
James Shubin
5d664855de lang: interfaces, funcs: Implement fmt.Stringer for functions
This adds the requirement that all function implementations provider a
String() string method so that these can be used as vertices in the
pgraph library. If we eventually move to generics for the pgraph DAG,
then this might not matter, but it's not bad that these have names
either.
2023-03-03 14:12:09 -05:00
James Shubin
8366cf0873 pgraph: Add a named error for detected cycles
This makes it useful to find when a topological sort failed.
2023-02-25 23:20:05 -05:00
James Shubin
a41789a746 examples, lang: funcs: core: example: Update vumeter
This patch moves to use the sox package instead of arecord for getting
microphone data, and it also validates that both sox and rec and
installed. We also add a standalone example.
2023-02-08 12:14:11 -05:00
James Shubin
cde3251dd8 test: shell: Disable flaky network test 2022-11-11 22:42:29 -05:00
James Shubin
7c394bf735 lang: Move the Edge struct into the interfaces package
This makes it consumable from more than one package and avoids future
cycles.
2022-11-11 20:28:22 -05:00
James Shubin
76e0345609 pgraph: Make some functions variadic for consistency
This will make it more useful when consuming these functions in a Txn
API which might be more convenient as a big one-liner.
2022-11-11 20:20:14 -05:00
James Shubin
d8820fa185 make: Improve crossbuild docs 2022-10-24 02:38:38 -04:00
James Shubin
b6502693e4 make, golang: Work around new go.mod issues
This updates the versions of virtually all our deps. Ever since golang
switched to go.mod we were unable to do so due to blockages with the
retract mechanism that one dep had used. Here's what seems like a
workaround for now. Suggestions from expert welcome.
2022-10-17 19:36:12 -04:00
James Shubin
f7e5402966 make: Remove missing clean target
This is no longer present.
2022-09-12 14:27:39 -04:00
James Shubin
1e6a825412 lang: funcs: Add cancel methods in a different way
Previously go vet found:
"this return statement may be reached without using the cancel var
defined on line..."
2022-09-12 14:11:25 -04:00
Samuel Gélineau
c23065aacd lang: funcs: Add SystemFunc
Runs a string as a shell command, then produces each line from stdout.
2022-09-11 21:58:39 -04:00
James Shubin
04f5ba67a2 lang: Small build fixes 2022-09-11 21:43:57 -04:00
James Shubin
b87fa6715b misc: Exclude more directories from ack matching
This is almost always what we want.
2022-09-11 20:55:37 -04:00
James Shubin
f6f3298e03 misc: Use go install instead of go get for build binaries 2022-09-11 20:55:37 -04:00
James Shubin
6bfd781947 lang: Replace the go-bindata usage with embed
This doesn't let us have nested mcl at the moment, but we could improve
on this with an embed API for each package. For now this makes building
the project easier.
2022-09-11 20:55:37 -04:00
James Shubin
aff6331211 misc, docs: Move to golang 1.18
Hopefully this fixes some yucky things.
2022-09-11 20:55:37 -04:00
James Shubin
d547c39a16 misc, docs: Move to golang 1.17
There are a few small source fixes and other tidbits to move past 1.16,
but we can do those later.
2022-08-16 19:30:20 -04:00
James Shubin
3cea422365 legal: Happy 2022 everyone...
Done with:

ack '2021+' -l | xargs sed -i -e 's/2021+/2022+/g'

Checked manually with:

git add -p

Hello to future James from 2023, and Happy Hacking!
2022-08-05 23:06:27 -04:00
James Shubin
ac39606386 lang: Misc changes from an old feature branch 2022-08-04 14:49:24 -04:00
James Shubin
12ae44d563 lang: funcs: core: fmt: Add important comment to printf 2022-08-04 14:36:56 -04:00
James Shubin
57b37d9005 pgraph: Misc cleanups and additions that I had kicking around
These changes had been sitting in a feature branch for a while. Push
them up.
2022-08-04 14:36:56 -04:00
James Shubin
9d5cc07567 lang: funcs: core: iter: Add map iterator function part3
This flattens the type unification of the map function so that the
solver has more to work with. It's possible that some scenarios might
solve faster, or without recursion, after this improvement.
2022-08-04 14:24:38 -04:00
James Shubin
75d4d767c6 lang: funcs: core: iter: Add map iterator function part2
This adds the Unify method to our map function and also switches the arg
order because I decided it would look nicer. Completely untested.
2022-08-04 14:24:05 -04:00
James Shubin
0be4b86230 lang: funcs: core: iter: Add map iterator function
Sadly this doesn't all work yet, but the tests and xmap function are
approximately correct. Eventually we add filter and reduce too!
2022-08-04 14:23:16 -04:00
James Shubin
784d15b012 all: Misc housekeeping for new golang versions 2022-08-04 14:16:33 -04:00
James Shubin
00f6045b12 util: safepath: Add a new Dir method and tests
This adds a new helper method. It should be considered for other types
as well.
2022-05-10 23:44:48 -04:00
Jef Masereel
b26f842de1 engine: resources: Add hetzner:vm resource
Hetzner cloud resource using hcloud-go. Requires polling via Meta:poll param. This first commit provides a stable vm resource with support for the basic functions of creating, deleting and updating a live server instance. SSH key handling does still require some attention to make sure checkapply can detect and update live changes to the specified keylist. A dedicated hetzner:sshkeys resource might be in order to make sure the keyset is handled correctly if there are multiple hetzner:vm resources running under the same Hetzner project. All remarks for future improvements are indicated with a TODO prefix
2022-04-20 14:24:25 -04:00
James Shubin
0ab2406db9 engine: Pass through the program version
We forgot this in a few places.
2022-02-17 13:34:26 -05:00
Joe Groocock
bf7e45439b engine: util: Skip unexported fields in struct field mapping
Unexported fields should be ignored when mapping structs from Golang to
mcl, as this avoids issues where certain structs cannot be used in mcl
representation due to their structure that may be incompatible with
conversion. Disallowing private fields allows those to still be included
in the struct but ignored when mapping the types, allowing the mcl
representation of the type to stil be valid

Resolves: https://github.com/purpleidea/mgmt/issues/684
Signed-off-by: Joe Groocock <me@frebib.net>
2021-12-01 22:06:06 +00:00
dantefromhell
0652273fe1 misc: Add systemd service file to release packages
According to [1] "/usr/lib/systemd/system/" is where systemd files from installed packages are stored in Arch Linux

Big thanks to @jordansissel for the in-depth explanation of this PRs effects:
> This PR replaces --prefix with the fpm dir package type's feature "path mapping" which lets you map a local file to a destination file when creating the package using this syntax: localpath=destinationpath

[1] https://wiki.archlinux.org/title/Systemd#Writing_unit_files
2021-11-09 21:42:50 +00:00
James Shubin
5927a54208 docs: Improve autogenerate godoc
There were a bunch of packages that weren't well documented. With the
recent split up of the lang package, I figured it would be more helpful
for new contributors who want to learn the structure of the project.
2021-10-26 00:12:18 -04:00
James Shubin
b46db59948 readme: Update to new godoc link
Thanks to Drew DeVault for hosting this.
2021-10-22 17:44:08 -04:00
James Shubin
23b5a4729f lang: Split lang package out into many subpackages
This is a giant refactor to split the giant lang package into many
subpackages. The most difficult piece was figuring out how to extract
the extra ast structs into their own package, because they needed to
call two functions which also needed to import the ast.

The solution was to separate out those functions into their own
packages, and to pass them into the ast at the root when they're needed,
and to let the relevant ast portions call a handle.

This isn't terribly ugly because we already had a giant data struct
woven through the ast.

The bad part is rebasing any WIP work on top of this.
2021-10-21 04:13:04 -04:00
James Shubin
8ae47bd490 test: Disable flaky shell tests
We need to improve and add these things back eventually, but for now I
don't want intermittent failures.
2021-10-21 01:17:52 -04:00
James Shubin
1796d20399 lang: Split the download code out into its own package
More cleanups.
2021-10-21 00:05:37 -04:00
James Shubin
5ac2447b85 lang: Split out the GAPI code for the lang frontend
This can go in a separate package.
2021-10-21 00:03:07 -04:00
James Shubin
db445c3a8e lang: Move the inputs logic into a separate package 2021-10-20 18:13:46 -04:00
James Shubin
de2914978d lang: Move interpret function to a separate package 2021-10-20 17:50:21 -04:00
James Shubin
09812a7bfc lang: Rename prog struct field to body
This is more consistent and avoids the future prog.prog stuttering.
2021-10-20 16:24:14 -04:00
James Shubin
2eb3b541f4 lang: Fix typo 2021-10-20 15:32:43 -04:00
James Shubin
e9791ff92c lang: funcs: core: fmt: Add variant verb for printf
There's no reason we can't support a %v variant verb. Of course it makes
type unification more difficult, and certain uses of this will produce
unsolvable situations, but it's useful for debugging, and fun to have.
2021-10-11 00:36:29 -04:00
James Shubin
88516546fa lib: Move to go modules and shed a few tears
The old system with vendor/ and git submodules worked great,
unfortunately FUD around git submodules seemed to scare people away and
golang moved to a go.mod system that adds a new lock file format instead
of using the built-in git version. It's now almost impossible to use
modern golang without this, so we've switched.

So much for the golang compatibility promise-- turns out it doesn't
apply to the useful parts that I actually care about like this.

Thanks to frebib for his incredibly valuable contributions to this
patch. This snide commit message is mine alone.

This patch also mixes in some changes due to legacy golang as we've also
bumped the minimum version to 1.16 in the docs and tests.

Lastly, we had to disable some tests and fix up a few other misc things
to get this passing. We've definitely hot bugs in the go.mod system, and
our Makefile tries to workaround those.
2021-10-05 08:34:51 -04:00
James Shubin
9c75c55fa4 lib: Update the help text
Give some longer descriptions so they show up nicely for the user.
2021-09-29 02:11:19 -04:00
Joe Groocock
b9741e87bd lang: interpolate: Fix string interpolation of dollar symbols
Dollar symbols were failing to parse when not followed by a non-brace,
non-dollar, non-EOF token and causing expected tests to fail. This
simplifies the rules to allow the remaining tests to succeed.

Fix and reinstate the final few failing tests, and add another.

Allow any escape sequence to be matched so that invalid sequences
produce a meaningful error message instead of a generic "cannot parse":

    ast: interpolate: interpolating: V: \?
    unhandled escape sequence token: \?

Tidy the related Makefile rule for generating the ragel parser.

Signed-off-by: Joe Groocock <me@frebib.net>
2021-09-28 21:40:49 +00:00
James Shubin
c555478b54 engine, lang: Misc fixes for golang lint 2021-08-09 16:55:31 -04:00
Joe Groocock
3718372288 docs: Provide Libera.Chat webchat links over ircs URIs
Only a few days after updating the documentation[1] following the move
to libera.chat, the webchat client was added, mirroring the behaviour of
the documentation prior to the change. Replace the ircs:// links with
clickable URLs to a usable browser chat client, which is more ideal for
beginners. Advanced users will know what to do to connect using their
external client as normal

[1]: 7d7e225823

Signed-off-by: Joe Groocock <me@frebib.net>
2021-07-12 20:57:44 +00:00
James Shubin
390b41bc26 test: Add small test for weird bash spacing 2021-06-21 18:28:05 -04:00
James Shubin
530c5a64fb vendor: Pin version of consul until we're on golang 1.16
The builds broke because the consul dependency now requires golang 1.16,
so let's pin it for now.
2021-06-20 21:53:01 -04:00
Matthew Lesko-Krleza
d285aaedc9 test: Add a test for AddEdge 2021-05-30 21:08:37 -04:00
James Shubin
453fe18d7f lang: Move the Arg type into the common interface package
This lets it get used in multiple places.
2021-05-30 17:59:50 -04:00
James Shubin
5fae5cd308 lang: Fix grammar typos
Woops!
2021-05-30 17:16:49 -04:00
Joe Groocock
7d7e225823 docs: Update IRC links to Libera.Chat
#mgmtconfig has moved to Libera.Chat as the primary channel for IRC
communications. Update the documentation to reflect this.
Libera.Chat doesn't provide a first-party web portal but does recommend
a few in the linked documentation on the website. As there is no
suitable replacement for webchat.freenode.net, link to the "Choosing an
IRC client" page instead.

Signed-off-by: Joe Groocock <me@frebib.net>
2021-05-27 08:50:24 +01:00
viq
19f404799d test: Fix awk usage for printing refs
This should print more fields than just 3rd one if they're present.
2021-05-24 23:49:19 +02:00
James Shubin
3e4652dca3 lang: funcs: Fix structlookup unification bug
We had mapped the field type to a dummy type instead of to T2 the return
type. Fixed now and added some tests.

This broke the unification for the load function lookups.
2021-05-23 22:52:50 -04:00
James Shubin
45b08de874 docs: Update the function guide
Hopefully this makes it easier for new function authors to get going
faster!
2021-05-23 20:21:14 -04:00
James Shubin
310e26dda9 lang: Switch over to the new PolyFunc interface
This isn't perfect yet, but we're trying to do this incrementally, and
merge whatever we can as early as possible.

During this work, I realized that the Simplify method of the exclusive
could probably be improved, and possibly receive a better signature.
This work will have to happen later.
2021-05-23 20:03:10 -04:00
James Shubin
f4eb54b835 lang: funcs: Add more invariants to contains func
This adds even more invariants to contains that I might have missed. It
may be redundant, or it may help. It also adds some tests.
2021-05-23 20:03:10 -04:00
James Shubin
3968c12947 lang: funcs: simplepoly: Support variant's in func definitions
This adds support for variant types in the simple poly definitions. It
is recommended that you avoid using these as much as possible, because
they're a bit harder for the type unification to solve for them. The way
this works is that these functions look at the available input types and
then generate a (recursive) set of invariants which might hold true. It
filters out any impossible ones, which is where this variant matching is
done. It's less likely that you'll get a solution with this mechanism,
but it is possible.
2021-05-23 20:03:10 -04:00
James Shubin
21c97d255f lang: funcs: simple: Check for function signatures
Make sure that we actually get function types here. This is just an
extra safety check.
2021-05-23 20:03:10 -04:00
James Shubin
eb1053607a lang: funcs: simple: Check for variant signatures
This adds a safety check in case someone sneaks in a variant type in the
simple function signature. These might be sneaky to detect, and it's
simpler to catch them right here.

From a design point of view, we might consider actually permitting
these, like we did with the simple poly API, but it's probably better
for them to get implemented in that API instead (if we decide to allow
this long-term) and keep this simple API very simple.
2021-05-23 20:03:10 -04:00
James Shubin
de7198e9dc lang: funcs: Check for functions that haven't been migrated
All polymorphic functions should use the new API, at least until we
either implement a compat wrapper. But it's probably best if we get rid
of the old API as soon as we make all this type unification work
properly.
2021-05-23 20:03:10 -04:00
James Shubin
0f30f47249 lang: funcs: core: world: Add unification to schedule return expr
This adds a sneaky unification between the expression of the function
return value in the unification. I am not entirely sure how often this
will get used, but it could be valuable in the right instance if this
isn't already learned through other sources. I'm fairly confident that
it isn't incorrect, so in the worst case scenario it's redundant
information for the unification solver.

This is being added as a separate commit so that it's obvious how this
type of unification invariant can be applied.
2021-05-23 20:03:10 -04:00
James Shubin
6b2ad8ebc8 lang: funcs: core: world: Add Unify method for schedule function
We should probably add some tests for this function because it once had
type unification ghosts, and while adding this new API method, I somehow
hit some temporary new ghosts that have since been killed.
2021-05-23 20:03:10 -04:00
James Shubin
1f302144ef lang: funcs: core: world: Move schedule func arg names to a const
This is a bit safer and cleaner.
2021-05-23 20:03:10 -04:00
James Shubin
d04c7a6ae4 lang: funcs: Add Unify method for history function
This could use some tests.
2021-05-23 20:03:10 -04:00
James Shubin
9ca2cda8c7 lang: funcs: core: Add more invariants to template func
This adds even more invariants to template that I might have missed. It
may be redundant, or it may help.
2021-05-23 20:03:10 -04:00
James Shubin
1fd06ecbf9 lang: funcs: core: fmt: Add more invariants to printf
This adds even more invariants to print that I might have missed. It may
be redundant, or it may help.
2021-05-23 20:03:10 -04:00
James Shubin
97baad4cb1 lang: funcs: Add Unify method for maplookup function
This also adds a few tests.
2021-05-23 20:03:10 -04:00
James Shubin
fbd93ecf0d lang: funcs: Add Unify method for structlookup function
This also adds a few tests.
2021-05-23 20:03:10 -04:00
James Shubin
e941ccea92 lang: funcs: Add Unify method for the simplepoly API
This is an implementation of the Unify approach for the simplepoly
function API, which wraps the full function API. It is unique in that a
lot of different functions use it, and it is easy to build functions
with it. It needs to use exclusives to represent the different options,
but at least it filters out any that aren't viable.

The Unify implementation here is fairly similar to the patterns in the
template() function.

To improve the filtering, it would be excellent if we could examine the
return type in `solved` somehow (if it is known) and use that to trim
our list of exclusives down even further! The smaller exclusives are,
the faster everything in the solver can run.
2021-05-23 20:03:10 -04:00
James Shubin
d692483bc3 lang: funcs: Add Unify method for operator function
This is an implementation of the Unify approach for the operator
function. It is unique in that it is a wrapper around the simple
operator function API.

To improve the filtering, it would be excellent if we could examine the
return type in `solved` somehow (if it is known) and use that to trim
our list of exclusives down even further! The smaller exclusives are,
the faster everything in the solver can run.
2021-05-23 20:03:10 -04:00
James Shubin
95cfbd0fff lang: funcs: Ensure that Info sig's are invalid if not built yet
In case something in the type unification tries to speculatively call
Info before it's ready to produce a valid sig, make sure we only return
a definitive answer (non-nil, and no variant types) once we've
conclusively finished defining the signature.
2021-05-23 20:03:10 -04:00
James Shubin
b3d1ed9e65 lang: funcs: core: math: Add a fortytwo function
This is mainly meant as a useful test case, but might as well have it be
fun too. As an aside, it taught me a surprising result about the %v verb
in printf, and we'll have to decide if it's an issue we care about.

https://github.com/golang/go/issues/46118

The interesting thing about this method is that it uses the simplepoly
API but has no input args-- only the output types are different. If it
had identical types in the input args, that might also have been
interesting, but it's more rare to have none. Hopefully this exercises
our type unification logic.
2021-05-12 03:30:25 -04:00
Joe Groocock
fe2b8c9fee engine: resources: exec: AutoEdge to User/Group/File
Fixes https://github.com/purpleidea/mgmt/issues/221

Signed-off-by: Joe Groocock <me@frebib.net>
2021-05-11 16:51:02 -04:00
James Shubin
2d7deef4e2 lang: unification: Don't stall the solver over generators
If we have a solution, and all that remains are generators, then feel
free to remove them and win.
2021-05-11 05:23:00 -04:00
James Shubin
b4a70b02e3 lang: funcs: Add Unify method for contains function
This is an implementation of the Unify approach for the contains
function. It is unique in that its generator invariant can recursively
generate a new generator invariant once.
2021-05-11 04:41:32 -04:00
James Shubin
c5c2364ed4 lang: funcs: core: fmt: Add an additional invariant to printf
This adds an invariant for printf that I might have missed. It may be
redundant, or it may help.
2021-05-11 03:23:33 -04:00
James Shubin
efcc4291a3 lang: funcs: core: Add Unify method for template function
This is an implementation of the Unify approach for the template
function.
2021-05-11 03:22:27 -04:00
James Shubin
6ea6ee264d lang: Add new unification rules for functions
This is meant as an incremental step into the new unification. Hopefully
it doesn't break anything and that we can rip out the old polymorphisms
work soon.
2021-05-11 02:52:35 -04:00
James Shubin
2865ba7632 lang: unification: Improve our simple solver
This removed a bug in the InvariantCall stuff, and also hopefully made
it more robust to actually solving when it had a solution.
2021-05-11 02:47:24 -04:00
James Shubin
2bed668d31 lang: interfaces: Small fixups to make unification work for now
This is all hacks until it works. Sorry that I am not a type unification
expert. If you are, please send us some patches =D
2021-05-11 01:31:10 -04:00
James Shubin
9dc24860f3 lang: interfaces: Add a new poly func interface
This new interface is subject to change and will probably be renamed if
we decide to keep it.
2021-05-11 00:45:25 -04:00
James Shubin
f01377b3bc lang: funcs: core: fmt: Add Unify method for printf
This is an implementation of the Unify approach for the printf function.
2021-05-11 00:33:50 -04:00
Joe Groocock
7443dfac4c misc: Run apt update before installing packages
Sometimes the package repo may be out of date and installing required
packages can return 404 because the version in the stale database has
been removed.

Signed-off-by: Joe Groocock <me@frebib.net>
2021-05-09 19:22:40 +01:00
James Shubin
e6408e187c lang: Rename old fail5 and fail6 variables
Last of the numbered error scenario cleanups...
2021-05-08 05:19:36 -04:00
James Shubin
a02d282d3e lang: Rename old fail8 variable to failInterpolate
Eight... Interpolate?

Cleanups...
2021-05-08 04:30:17 -04:00
James Shubin
f778f53744 lang: Rename old fail9 variable to failInit
Cleanups...
2021-05-08 04:25:56 -04:00
James Shubin
95ea93564e lang: Rename old fail4 variable to failGraph
Cleanups...
2021-05-08 04:20:05 -04:00
James Shubin
d51029e86c lang: Rename old fail3 variable to failUnify
Cleanups...
2021-05-08 04:17:21 -04:00
James Shubin
1016699c94 lang: Rename old fail2 variable to failSetScope
Cleanups...
2021-05-08 04:13:33 -04:00
James Shubin
63f63955e7 lang: Replace numbered errors with named ones
This makes the tests easier to read and modify without having out of
order numbers. When writing the tests, you'll remember more easily which
section you're erroring in too!
2021-05-07 23:41:00 -04:00
James Shubin
37be9fda9f lang: Catch duplicate resource fields or meta entries statically
This teaches the compiler to catch entries with duplicate fields, and
duplicate meta entries, because it could be ambiguous to determine which
should take precedence. For example, if you specified `content` to a
file resource twice, this should error. This is known statically, so we
can catch it. If you specified two `Meta:noop` entries, this can also be
caught.

The interesting part happens when you specify one `Meta:noop` entry, and
one `Meta` entry which happens to contain a noop field in the struct.
For this, we actually have to wait until type unification is finished,
and catch the error there. This is because after type unification we
will know the precise type of the struct being passed to `Meta`, and so
we can look at its field names, even if their values aren't yet known
because the graph hasn't run yet.
2021-05-07 23:41:00 -04:00
James Shubin
0756133a7e lang: Add a named error for catching errors in test on Init
This makes it so that we can catch errors that happen in Init. We also
name the errors so that number sequence doesn't matter.
2021-05-07 23:14:49 -04:00
Joe Groocock
83c5ab318b lang: types: Clear map/list types during Into()
Map and list types are now unconditionally initialised during an Into()
call to ensure that the only data within them after the operation is
that added by the Into() function.

Prior to this change, map/list types would likely not be cleared prior
to the data being inserted into them with a few exceptions. Nil
pointers or maps/lists that were not sufficient in capacity would be
reinitialised and used to replace the existing backing data store. In
some cases this wouldn't occur meaning any residual data existing in the
container before the Into() call could persist after the data copy
completes. This behaviour is wildly inconsistent and not ideal in the
vast majority of cases. It should be assumed that the Into() call will
preserve nothing and always produce a consistent and deterministic
output.

Signed-off-by: Joe Groocock <me@frebib.net>
2021-05-05 10:41:48 +01:00
Joe Groocock
0c28957016 lang: funcs: Funcs that never load are fatal
If there is a programming error in any func Stream() implementation then
the node could never output anything, causing the engine to hang
indefinitely waiting for an initial value that will never come,

Nodes keep track of whether they are loaded, so testing for this
occurence is pretty simple. Any nodes that do not return output at least
once before they close their output channel can be considered a fatal
error on which the engine will exit.

Signed-off-by: Joe Groocock <me@frebib.net>
2021-05-04 11:39:34 -04:00
James Shubin
959084040d lang: Don't block the engine for empty values
If the user passed an empty list or map, we should send that and not
block. This also includes a simple test to ensure this keeps working.
2021-05-04 11:27:58 -04:00
James Shubin
8a428c6936 lang: Add a test timeout to catch blocked cases
This should catch any blocked tests and report an error. The timeout is
arbitrary.
2021-05-04 11:26:17 -04:00
Joe Groocock
48da23226c project: Add frebib to AUTHORS
Signed-off-by: Joe Groocock <me@frebib.net>
2021-05-04 14:16:54 +01:00
James Shubin
5f0c6e5102 lang: types: Add extra ValueOf test
Just to double check weird behaviours of the golang reflect lib.
2021-05-04 06:26:33 -04:00
Joe Groocock
29f1c6f50e lang: types: Fix ValueOf() panic with nil pointer values
Some forms of reflect.Value can cause ValueOf() to panic when there is a
nil pointer somewhere within the reflect.Value, whether that be a
container type like a struct, list or map, or just a raw nil pointer.

In these cases, ValueOf() attempted to dereference the pointer without
ever checking if it was nil. mgmt lang doesn't have pointers of any
kind, so these Golang values cannot be represented in mcl types in the
current form so return a helpful error to the user.

Signed-off-by: Joe Groocock <me@frebib.net>
2021-05-04 06:26:23 -04:00
James Shubin
4d187419ac engine: Small typo/cleanups in autogrouping code 2021-05-04 05:30:27 -04:00
James Shubin
58998f9cab engine: Transform the send/recv init functions into helpers
Since we'll want to use them elsewhere, we should make these helper
functions. It also makes the code look a lot neater. Unfortunately, it
adds a bit more indirection, but this isn't a critical flaw here.
2021-05-04 05:30:27 -04:00
James Shubin
cdc5ca8854 util: Add a simple log wrapper for io.Writer
This lets logger interfaces that use io.Writer get met by our logging
interface.
2021-05-04 05:27:07 -04:00
James Shubin
44e1e41266 lang: types: Improve documentation for ValueOf functions
A reminder that nil's in golang don't map to anything in mcl.
2021-05-04 05:27:07 -04:00
James Shubin
33fda8605a lang: types: Add new ValueOf tests
Hopefully this makes all of this a bit more obvious.
2021-05-04 04:27:33 -04:00
Joe Groocock
5f9ed69299 misc: Replace go-bindata with maintained fork
As per [1] go-bindata was removed from GitHub and later replaced by the
community. jteeuwen/go-bindata has since been archived to represent this
state and now most communities use kevinburke/go-bindata instead as it
is more actively maintained.

[1]: https://github.com/jteeuwen/go-bindata/issues/5

Signed-off-by: Joe Groocock <me@frebib.net>
2021-05-04 04:10:05 -04:00
Joe Groocock
7f1baea3b0 engine: resources: docker: Replace deprecated NewClient() with NewClientWithOpts()
docker/client.NewClient() is deprecated in favour of NewClientWithOpts()
which takes a series of client.Opt functions to configure the API
client. As mgmt only passes the API version through, this simplifies the
NewClient() calls.

Signed-off-by: Joe Groocock <me@frebib.net>
2021-05-02 10:39:04 +01:00
James Shubin
f75026e4b2 lang: unification: Teach the solver about new invariants
This extends our simple solver so that it can understand the new
invariants. For the Value holding invariants, it doesn't do much-- those
are expected to be used within the execution of the GeneratorInvariant
so they are passed through untouched. For the GeneratorInvariant it must
actually try running this periodically to see if it produces new
invariants that are helpful for the whole solution. Since this could get
expensive quickly, the logic is to only try running these once we've
entered steady state, but before we've tried to reach for the
ExclusiveInvariant's. The exclusives are the most expensive so we run
these last, and the generators are run late because they won't usually
produce anything helpful unless some of the basic solving has already
happened. If they could produce useful things right away, then there
wouldn't be a need for them!
2021-05-02 00:52:57 -04:00
James Shubin
ce7a1a9c67 lang: Add a CallFuncArgsValueInvariant invariant
This is a new invariant that I realized might be useful. It's not
guaranteed that it will be used or useful, but I wanted to get it out of
my WIP branch, to keep that work cleaner.
2021-05-02 00:52:57 -04:00
James Shubin
a62056fb19 lang: Add a GeneratorInvariant invariant
This is a new invariant that I realized might be useful. It's not
guaranteed that it will be used or useful, but I wanted to get it out of
my WIP branch, to keep that work cleaner.
2021-05-02 00:52:57 -04:00
James Shubin
f3434a8155 lang: Add a ValueInvariant invariant
This is a new invariant that I realized might be useful. It's not
guaranteed that it will be used or useful, but I wanted to get it out of
my WIP branch, to keep that work cleaner.
2021-05-02 00:52:57 -04:00
James Shubin
4e023ef517 lang: Move the ExprAny to the interfaces package
Having this special "placeholder" interface is useful for more than one
package.
2021-05-02 00:52:57 -04:00
James Shubin
97b80cb930 lang: unification: Move the InvariantSolution struct
We are just relocating this in the file for consistency.
2021-05-02 00:52:57 -04:00
James Shubin
525b4e6a53 lang: Move core unification structs into shared interfaces package
We should probably move these into the central interfaces package so
that these can be used from multiple places. They don't have any
dependencies, and it doesn't make sense to have the solver code mixed in
to the same package. Overall the interface being implemented here could
probably be improved, but that's a project for another day.
2021-05-02 00:52:57 -04:00
James Shubin
054eaf65b8 util: safepath: Add a new safe path helper library
This is a new path manipulation library that is designed to be safer
than using simple strings for everything. It is more work to use, but it
can help you keep track of the different path types.

It has been sitting unused in a git branch for too long, and I figured
it should see the light of day in case someone wants to start using it.
2021-05-02 00:52:57 -04:00
James Shubin
48fa796ab1 test: Disable failing test
Hit another intermittent failed test in GH CI.
2021-03-10 03:36:27 -05:00
Jean-Philippe Evrard
1873e022cc test: Add guard when no commit needs testing
Without this patch, github actions fail.

It's a temporary workaround until [1] is done.

[1]: https://github.com/purpleidea/mgmt/issues/643
2021-03-03 10:49:22 +01:00
James Shubin
35a8062b58 test: Disable travis IRC notifications for now
One of our contributors is unusually annoyed by them, and it's important
to keep your contributors happy!
2021-03-03 04:28:53 -05:00
Jean-Philippe Evrard
636248ad67 test: Ensure branches are also testable
test-commit-message runs on PR, but also on push in other branches
which aren't PRs. We need to test those too.

This is fixed by ensuring the same kind of behaviour than travis CI:
When a patch is put on a branch, it's using the branchname for
testing [1].

[1]:
https://docs-staging.travis-ci.com/user/environment-variables/#default-environment-variables
2021-03-03 09:52:49 +01:00
Jean-Philippe Evrard
4511c54fad test: Ensure github CI tests commit messages
Without this patch, the travis var is empty, and we just pass.
This is a problem, as we are using github CI nowadays.

This should fix it.
2021-03-03 02:47:30 -05:00
James Shubin
7f3970541b test: Skip more tests
I think some of these fail due to shared environments and noisy
neighbours in github. We'll have to fix that eventually or test
elsewhere.
2021-03-02 13:41:00 -05:00
James Shubin
4040f4d151 test: Skip yet another intermittent test
We shall not have intermittent tests!
2021-03-02 12:48:44 -05:00
James Shubin
887d374c53 lang: funcs: Catch simple function api usage without types
In case a programmer makes a mistake and passes in a function using the
simple function API without a type or even without the entire value,
we'll now return a sensible error message and panic in init() instead of
requiring a test to catch this alone.
2021-02-28 22:50:51 -05:00
James Shubin
be4b87155d test: Skip another intermittent test
I think this might be related to multiple jobs running at the same time
on the same host. Not sure though.
2021-02-24 04:00:34 -05:00
James Shubin
b987a7da4c examples: tftp: Fix missing error checking in example 2021-02-20 13:31:45 -05:00
James Shubin
7153fe5ad2 test: Skip intermittent tests
It would be great to fix some rare races or debug what's wrong in CI,
but for now let's get rid of these fails so that we can get better data
for when we break something more serious. We'll need to revisit all of
this for sure.
2021-02-19 21:17:57 -05:00
James Shubin
ccd8ba44d9 test: Exclude generated ragel parser from golint 2021-02-17 22:01:34 -05:00
James Shubin
e7ef0f7a6c test: Set default column size if $TERM env var isn't set
Seems Github actions breaks or unsets this, leading to the errors:

tput: No value for $TERM and no -T specified
seq: missing operand
Try 'seq --help' for more information.

Hopefully this makes things a bit more robust.
2021-02-17 04:07:30 -05:00
James Shubin
400b58c0e9 lang: Improve string interpolation
The original string interpolation was based on hil which didn't allow
proper escaping, since they used a different escape pattern. Secondly,
the golang Unquote function didn't deal with the variable substitution,
which meant it had to be performed in a second step.

Most importantly, because we did this partial job in Unquote (the fact
that is strips the leading and trailing quotes tricked me into thinking
I was done with interpolation!) it was impossible to remedy the
remaining parts in a second pass with hil. Both operations needs to be
done in a single step. This is logical when you aren't tunnel visioned.

This patch replaces both of these so that string interpolation works
properly. This removes the ability to allow inline function calls in a
string, however this was an incidental feature, and it's not clear that
having it is a good idea. It also requires you wrap the var name with
curly braces. (They are not optional.)

This comes with a load of tests, but I think I got some of it wrong,
since I'm quite new at ragel. If you find something, please say so =D In
any case, this is much better than the original hil implementation, and
easy for a new contributor to patch to make the necessary fixes.
2021-02-17 03:35:12 -05:00
James Shubin
5257496214 test: Make a few cosmetic changes and enable race testing 2021-02-17 02:45:47 -05:00
Jean-Philippe Evrard
e1bfe4a3ce test: Add GitHub Actions test support
Authored-By: Jean-Philippe Evrard <open-source@a.spamming.party>
Co-Authored-By: James Shubin <james@shubin.ca>
Signed-off-by: Joe Groocock <me@frebib.net>
2021-02-14 21:41:02 -05:00
James Shubin
f31cce8ec2 test: misc: Clarify golang wording 2021-02-14 21:39:01 -05:00
Joe Groocock
169ebfa72c test: make-deps: Add folds around tests and dep blocks
Improves readability of CI test output and hides away the complexity
when in most cases it is not required. Retain fold behaviour for both
Travis and GitHub Actions in case both are used in any capacity.

Signed-off-by: Joe Groocock <me@frebib.net>
2021-02-12 16:44:12 +00:00
Joe Groocock
7cace52ab5 test: prevent LinuxBrew in GitHub Actions CI
Ubuntu-latest in GitHub Actions provides linuxbrew, so the tests install
both the native Debian dependency packages, and also the linuxbrew
variants which is slower and entirely redundant.

Signed-off-by: Joe Groocock <me@frebib.net>
2021-02-12 16:44:12 +00:00
Joe Groocock
95b93c60d9 test: Invert negative bash assertions
In bash `-n` is `non zero length` which is the opposite of `-z` meaning
`zero length`. `-n` is semantically identical to `! -z` but `-n`.

Signed-off-by: Joe Groocock <me@frebib.net>
2021-02-12 16:44:12 +00:00
Joe Groocock
5af1dcb8b1 test: Add in_ci utility test function
in_ci checks for environment variables set by a selection of CI systems
and returns true if the test appears to be running in CI. Additionally
it can test for specific CI systems, and returns true if the CI system
is listed.

Deduplicate existing environment checks for Travis and Jenkins.

Signed-off-by: Joe Groocock <me@frebib.net>
2021-02-12 16:44:11 +00:00
Joe Groocock
6a61774fb7 docker: Bump docker dependencies, add containerd
These dependencies are maintained because the upstream repos bundle
vendor directories into the repos, which cause namespacing issues during
build. Git submodules don't strip the vendor directory whereas most
vendoring tools would.

Signed-off-by: Joe Groocock <me@frebib.net>
2021-02-09 21:14:34 +00:00
James Shubin
ccbaca24f1 gopath: Remove this unused directory
I had this symlink hack a long time ago. I don't think it's being used
anymore.
2021-02-07 21:23:49 -05:00
Joe Groocock
07b6048dc5 etcd: Bump etcd + friends to the latest upstream version
This allows dropping the pinned grpc-prometheus and grpc-gateway
libraries as git master works fine for now.

Signed-off-by: Joe Groocock <me@frebib.net>
2021-02-07 12:55:02 +00:00
Joe Groocock
60dd34d066 make: Drop support for Go 1.9 in make build
docs/development.md says the minimum required Golang version is 1.13 at
the time of writing.

Signed-off-by: Joe Groocock <me@frebib.net>
2021-02-07 00:08:51 -05:00
James Shubin
28451d1e14 lang: funcs: Fixup race in vumeter example
The vumeter example was written quickly and without much care. This
fixes a possible race (panic) and also removes the busy loop that wastes
CPU while we're waiting for the first value to come in.
2021-02-06 23:59:06 -05:00
Joe Groocock
db95b6381f examples: lang: Reinstate mcl as unification bug is fixed
Struct field names now correctly map based on their `lang` tags in Go
structs, so this example now works as originally intended.

Signed-off-by: Joe Groocock <me@frebib.net>
2021-02-06 16:57:01 +00:00
Joe Groocock
6b14c9bea4 lang: Map Go struct fields using lang struct tag
Converting a reflect.Type of KindStruct did not respect the `lang` tag
on struct fields incidating how fields from mcl structs should be mapped
even though resource field names did. This patch should allow structs
with mapped fields to be respected.

Signed-off-by: Joe Groocock <me@frebib.net>
2021-02-06 16:57:01 +00:00
Joe Groocock
742adc00fe lang: Convert StmtRes to engine.Res with types.Into()
Replace existing field-mapping code with calls to types.Into() to
reflect the mcl data into the Go resource struct with finer granularity
and accuracy, and less reliance on the magic reflect.Set() function.

One major advantage over reflect.Value.Set() is Into() allows tailoring
the data that is set, specifically when coercing mcl struct values into
Golang struct values, the fields can be appropriately mapped based on
the lang tag on the struct field. With reflect.Value.Set() this was not
at all possible as there was a contradiction of logic given the
following rules:

- mcl struct fields must have lowercase names
- Golang struct fields with lowercase names are unexported
- Golang reflection does not allow modifying unexported fields

Prior to this change, it was impossible to map an mcl inline struct in a
resource to the matched Golang counterpart, even if the lang tag was
present on the struct field. This can be demonstrated with the following
trivial example:

    test "name" {
        key => struct{
            name => "hello",
        },
    }

and the accompanying Golang resource definition:

    type TestRes struct {
        traits.Base
        traits.Edgeable

        Key struct {
            Name string `lang:"name"`
        } `lang:"key"`
    }

Due to the mismatch in field names in the embedded struct, the type
unifier failed and refused to match mcl 'name' to Go 'Name' due to the
missing mapping logic.

Signed-off-by: Joe Groocock <me@frebib.net>
2021-02-06 16:57:00 +00:00
Joe Groocock
52897cc16c lang: Implement Into() to set types.Value from reflect.Value
Into() mutates a given reflect.Value and sets the data represented by
the types.Value into the storage represented by the reflect.Value.

Into() is the opposite to ValueOf() which converts reflect.Value into
types.Value, and in theory they should be (almost) bijective with some
edge case exceptions where the conversion is lossy.

Simply, it replaces reflect.Value.Set() in the broad case, giving finer
control of how the reflect.Value is modified and how the data is set.
types.Value.Value() is now also a redundant function that achieves the
same outcome as Into(), but with less type specificity.

Signed-off-by: Joe Groocock <me@frebib.net>
2021-02-06 16:57:00 +00:00
Joe Groocock
c950568f1b lang: Move StructTag const into lang/types
This constant value is strongly tied to the language, and little to do
with the engine. Move the definition into the lang/types package to
prevent circular imports between lang/types and engine/util.

Signed-off-by: Joe Groocock <me@frebib.net>
2021-02-06 16:56:57 +00:00
Joe Groocock
845d7ff188 lang: Fix panic in lang/types/ValueOf() for Struct
Replace use of reflect.Value.Len() with NumField() which is intended to
return the number of fields in reflected Struct value.

Len should only be used for Array, Chan, Map, Slice and String types.

Add some trivial sanity check tests for ValueOf() for the simple and
complex container types.

Signed-off-by: Joe Groocock <me@frebib.net>
2021-01-31 23:10:16 +00:00
James Shubin
3bd8658da6 art: Use a transparent logo for dark themes
As helpful user frebib pointed out, our logo doesn't render nicely when
github users use the dark theme. This has already been complained about
by others in:

https://github.community/t/support-theme-context-for-images-in-light-vs-dark-mode/147981/2

For now, we'll switch to a transparent background.
2021-01-31 18:00:47 -05:00
James Shubin
336a38081a legal: Happy 2021 everyone...
Done with:

ack '2020+' -l | xargs sed -i -e 's/2020+/2021+/g'

Checked manually with:

git add -p

Hello to future James from 2022, and Happy Hacking!
2021-01-31 16:52:46 -05:00
Jean-Philippe Evrard
01c2131436 misc: Make-deps should assume go vet present
While the code comment says to check if go vet command is present,
it actually tests if go vet command returns properly.

This is a problem if go vet is _not_ returning 0 due to a
failure while go vet is present: it will try to install the
legacy go vet.

This fixes it by removing this block of code completely,
as we assume a golang version which contains it anyway.
2020-12-09 19:25:19 -05:00
Joe Groocock
c274231544 test: Fix implicit test fail in test-markdownlint
Fix the silent test failure by catching the uncaught error from
`command`, handling the failure gracefully.

    $ bash -x test/test-markdownlint.sh
    ...
    ++ command -v mdl
    + MDL=
    $ echo $?
    1

    $ bash -x test/test-markdownlint.sh
    ...
    ++ command -v mdl
    + MDL=
    + true
    + '[' -z '' ']'
    + fail_test 'The '\''mdl'\'' utility can'\''t be found.'
    + echo -e 'FAIL: The '\''mdl'\'' utility can'\''t be found.'
    FAIL: The 'mdl' utility can't be found.

Fix a couple of glaring shellcheck warnings and errors mostly
surrounding variable quoting.

Signed-off-by: Joe Groocock <me@frebib.net>
2020-12-08 15:51:52 +00:00
Joe Groocock
4a2864701c lang: Assert that 'metadata.yaml' is not parsed as raw mcl
Contradictory to expectations, `mgmt run lang metadata.yaml` would
produce an error similar to the following, which is likely unexpected
from the user perspective:

    2020-11-24 12:24:08.330968 I | cli: lang: lexing/parsing...
    2020-11-24 12:24:08.331153 I | run: error: cli parse error: could not generate AST: parser: `syntax error: unexpected DOT` @1:1

Produce a user-friendly warning instead, hinting with the supported
usage:

    2020-11-24 13:15:01.686659 I | run: error: cli parse error: could not activate an input parser: unexpected raw code 'metadata.yaml'. Did you mean './metadata.yaml'?

Signed-off-by: Joe Groocock <me@frebib.net>
2020-11-24 13:56:08 +00:00
James Shubin
76ede10e0a misc: Update path to go-fuzz
They moved it :/
2020-09-23 13:42:51 -04:00
Ahmed Al-Hulaibi
274e01bb75 misc, docs: Update minimum required golang version to 1.13 2020-09-23 12:34:54 -04:00
James Shubin
d75f763c99 misc, docs: Move to golang 1.12 2020-09-23 12:34:54 -04:00
Donald Bakong
5bc985663c docs: Add underscore issue to FAQ 2020-09-23 11:25:46 -04:00
James Shubin
df9e2e853f docs: Update state of remote execution and resource 2020-09-23 11:21:03 -04:00
Ivan Pejić
b4828a6f0a docs: Update FAQ to mention temp absence of remote 2020-09-23 11:15:44 -04:00
Donald Bakong
e99dd749a0 engine: resources: Fixing return bug 2020-09-23 11:10:38 -04:00
David Randall
10ce7178c0 misc: Incorrect dependency path for dvyukov/go-fuzz
Fix the dependency path for dvyukov/go-fuzz to github.com from golang.org.

Fixes #601
2020-06-07 15:05:46 -04:00
Donald Bakong
5c6a66eaf5 lang: funcs: Add cidr_to_ip function 2020-06-03 22:01:01 -04:00
Francois Rompre-Lanctot
36d30bc985 lang: funcs: Add macfmt function 2020-06-03 17:46:39 -04:00
Adam Sigal
a5152b82e9 engine: resources: exec: Add Env
Add functionality to specify environment variables in exec.
2020-04-19 20:30:43 -06:00
James Shubin
e9af8a2595 engine: resources: exec: Clean up error handling
Some quick fixes, this whole resource should be looked at for
discrepancies, since it was written very early.
2020-04-14 22:59:33 -04:00
Adam Sigal
84b5b60d49 engine: resources: exec: Fix typo
Typo in description of cwd field fixed.
2020-04-14 22:36:25 -04:00
James Shubin
8f60f42be3 engine: resources: Add http:server and http:file resources
This adds a new http server resource, as well as a http file resource
that is used to specify files to serve in that server. This allows you
to have an http server that is entirely server from memory without
needing files on disk.

It does this by using the autogrouping magic that is already available
in the engine.

The http resource is not meant to be a full-featured http server
replacement, and it might still be useful to use the venerable webserver
of your choice, however for integrated, pure-golang bootstrapping
environments, this might prove to be very useful.

It can be combined with the tftp and dhcp resources to build PXE setups
with mgmt!

This resource can be extended further to support an http:flag endpoint,
an http:ui endpoint, automatic edges, and more!
2020-04-11 03:23:04 -04:00
James Shubin
583344138a engine: resources: Add dhcp:server and dhcp:host resources
This adds a new dhcp server resource, as well as a dhcp host resource
used to specify the static mapping between mac address and ip address.
It also adds a simple, pure-golang example dhcp client which might make
testing easier.

The dhcp resource is not meant to be a full-featured dhcp server
replacement, and it might still be useful to use the venerable dhcpd,
however for integrated, pure-golang bootstrapping environments, this
might prove to be very useful.

It can be combined with the tftp resource to build PXE setups with mgmt.

This resource can be extended further to support a dhcp:range directive,
automatic edges, and more!
2020-04-11 02:45:18 -04:00
James Shubin
016d021d5a engine: resources: tftp: Improve validation and error messages
Just some small cleanups for our tftp resource. We also rename the
struct to make it consistent, since golint complains about similar
protocols when it is not all capitalized.
2020-04-11 02:45:18 -04:00
James Shubin
115dc4bfa4 engine: resources: net: Add an IP forward field
This adds an IP forward field (boolean) and improves the documentation.
2020-04-11 02:02:35 -04:00
James Shubin
5b83febb23 etcd: Log an error that wasn't getting seen
We would error when the address could not bind, but it was for
non-standard reasons, and we didn't see the specific reason why.
2020-04-11 02:02:35 -04:00
James Shubin
c9d5c50402 test: Improve our test for long lines
This now allows long URL's to start part way through a sentence instead
of requiring them to start on the beginning of a new line.
2020-04-03 01:15:57 -04:00
James Shubin
fc839d2983 vendor: Fix broken upstream hashicorp module
This module broke compat with old versions of golang. Vendor it until
we're at a minimum of golang 1.13.x everywhere.
2020-04-02 22:27:06 -04:00
James Shubin
3bce96bbd5 docs: Add new talk from cfgmgmtcamp 2020
Those first ten seconds of the video are awesome!
2020-02-29 19:40:26 -05:00
Patrick Meyer
6279be073b lang: Prevent struct types with duplicate field names
The previous fix for #591 in 70eecd5 didn't address all issues
concerning duplicate struct field names. It still crashed for inputs
like `$d []struct{x int; x float}`. Note the different types but
duplicate names.
2020-02-29 19:11:51 -05:00
Patrick Meyer
ea37132ce4 lang: Fix wrong go-fuzz bin name
I missed renaming this after moving the fuzz.go from the lang package
to its own fuzz package.
2020-02-29 05:28:38 +01:00
James Shubin
70eecd5289 lang: Prevent struct types with duplicate fields
Struct types with duplicate fields are invalid types and weren't caught
by the parser. This fixes the issue and adds some associated tests. It
also checks and tests for duplicate struct value field names.

As a technical side-note, this doesn't change the lang/types/ functions
to remove panics-- the signatures are simplified to make their use
simple, and we intentionally panic if they're used incorrectly. In this
case, one was being used without having previously validated the input.

Thanks to Patrick Meyer for finding this issue via fuzzing!
2020-02-27 18:52:02 -05:00
James Shubin
380d03257f misc, lang: Small fixups
Change some minor style issues.
2020-02-27 17:34:03 -05:00
Patrick Meyer
006de6da14 lang: Add fuzz target to lang Makefile 2020-02-26 12:07:00 +01:00
Kenneth Hoste
10aa80e8f5 docs: Add link to recording of James' FOSDEM 2020 talks 2020-02-17 14:23:20 -05:00
Felix Frank
013439af6d test: Make prometheus tests safer and more verbose 2020-02-16 18:38:58 -05:00
Felix Frank
3408961155 test: Fix syntax in the loadavg test 2020-02-16 18:38:43 -05:00
Francois Rompre-Lanctot
f3b4a8d055 engine: resources: Add a test case for resource owner check
This adds FileOwnerExpect as a new Step which allows validating if the
owner was set properly on a resource.
2020-02-10 22:17:42 -05:00
Francois Rompre-Lanctot
104af7e86f engine: resources: Fix typo 2020-02-03 22:57:13 -05:00
James Shubin
be39fbeff6 examples: lang: Update examples 2020-02-01 16:48:23 -05:00
James Shubin
4109045fa4 github: Add new needinfo tag 2020-01-29 11:38:53 -05:00
James Shubin
90fd8023dd lang, engine: Add a facility for resources to export constants
Since we focus on safety, it would be nice to reduce the chance of any
runtime errors if we made a typo for a resource parameter. With this
patch, each resource can export constants into the global namespace so
that typos would cause a compile error.

Of course in the future if we had a more advanced type system, then we
could support precise types for each individual resource param, but in
an attempt to keep things simple, we'll leave that for another day. It
would add complexity too if we ever wanted to store a parameter
externally.

Lastly, we might consider adding "special case" parsing so that directly
specified fields would parse intelligently. For example, we could allow:

	file "/tmp/hello" {
		state => exists,	# magic sugar!
	}

This isn't supported for now, but if it works after all the other parser
changes have been made, it might be something to consider.
2020-01-29 11:16:04 -05:00
James Shubin
f67ad9c061 test: Add a check for too long or badly reflowed docstrings
This ensures that docstring comments are wrapped to 80 chars. ffrank
seemed to be making this mistake far too often, and it's a silly thing
to look for manually. As it turns out, I've made it too, as have many
others. Now we have a test that checks for most cases. There are still a
few stray cases that aren't checked automatically, but this can be
improved upon if someone is motivated to do so.

Before anyone complains about the 80 character limit: this only checks
docstring comments, not source code length or inline source code
comments. There's no excuse for having docstrings that are badly
reflowed or over 80 chars, particularly if you have an automated test.
2020-01-25 04:43:33 -05:00
James Shubin
525e2bafee misc: Let sigtee work on older golang versions 2020-01-25 04:43:33 -05:00
James Shubin
b65a9abf8e misc: Add two scripts to help debug things
This adds two new helper scripts that are good for debugging mgmt. Just
build and add `sigtee` to your ~/bin/ along with filter-golang-stack.py
and mgmt_debug.sh and use the later to call mgmt as you would normally.

For example, I might do:

$ mgmt_debug.sh ./mgmt run --tmp-prefix lang examples/lang/hello0.mcl

And if I kill it with ^\ then I'll get a filtered trace at the end in my
$PAGER (which is assumed to be `less`) and this should make my life
easier.

As a cool bonus, this means we use bash, python, and golang all
together!
2020-01-13 01:07:00 -05:00
James Shubin
fec94aa53a engine, lang: Fix simple test failures
Two bugs sneaked in while pushing old stuff.
2020-01-12 19:35:11 -05:00
James Shubin
3d4b345728 examples: lang: Improve reverse example
It's cool to show just the mode changes.
2020-01-12 17:42:20 -05:00
James Shubin
579975f08d engine: graph: Don't error when state file is missing
For some reason we get errors when we try to remove a non-existent state
file. There's a slight possibility that it could be a bug we're working
around, but it's not clear that this is the case, and I think it's
possible that a state file could have gotten nuked by the user somehow,
although this was occurring "naturally" when running reverse1.mcl so
let's keep that working for now.
2020-01-12 16:41:09 -05:00
James Shubin
3707b39fef engine: graph: Improve comments
Clarify that we're referring to cycles in the graph, since it needs to
be a DAG.
2020-01-12 16:39:32 -05:00
James Shubin
f07387225b engine: resources: Log more info about tftp errors
This helps for debugging this kind of issue:
https://github.com/pin/tftp/issues/41#issuecomment-570744056
2020-01-03 20:42:34 -05:00
James Shubin
2648fb1bb1 legal: Happy 2020 everyone...
Done with:

ack '2019+' -l | xargs sed -i -e 's/2019+/2020+/g'

Checked manually with:

git add -p

Hello to future James from 2021, and Happy Hacking!
2020-01-03 20:08:37 -05:00
James Shubin
d34715b4ba engine: resources: pippet: Cleanup and proper wrapping
Felix, please configure your editor to wrap at 80 chars and/or help us
write a test for this please =D
2020-01-03 01:20:00 -05:00
Felix Frank
63af50bf98 engine: resources: pippet: Initial implementation for new resource type
The pippet resource implements faster integration of Puppet resources
in mgmt at runtime, by piping synchronization commands to a Puppet
process that keeps running alongside mgmt. This avoids huge overhead
through launching a Puppet process for each operation on a resource
that is delegated to Puppet.
2020-01-03 01:19:37 -05:00
James Shubin
456550c1d4 engine: resources: docker: Make a few fixups
Here are a few fixups to the docker resources. All miscellaneous stuff,
nothing major.
2020-01-03 00:53:20 -05:00
Jonathan Gold
8174b88ec3 engine: resources: docker: Add sensible defaults 2020-01-03 00:30:01 -05:00
Jonathan Gold
3233973748 engine: resources: docker: Add AutoEdges between container and image 2020-01-03 00:30:01 -05:00
Jonathan Gold
bdfb1cf33e engine: resources: docker: Ensure image is specified for containers 2020-01-03 00:30:01 -05:00
Jonathan Gold
1c5fcd59e7 engine: resources: docker: Add a docker image resource 2020-01-03 00:30:01 -05:00
James Shubin
5cc960527e lang: funcs: Differentiate between empty and nil values
It would be good to differentiate between receiving an empty value or
not having received a value yet. This is similar to the previous commit.
2020-01-03 00:28:54 -05:00
James Shubin
762c53fb8d lang: funcs: Send empty values when appropriate
I seem to have forgotten to differentiate between the empty string and
no data because the zero value for the stored result was the empty
string. This turns it into a pointer so that we don't block the function
engine if a template or one of the other patched functions sends an
empty string as the first value.
2019-12-30 12:35:08 -05:00
James Shubin
ff20e67d07 lang: funcs: Improve template function
Template function should be able to be called with just one arg (no
input vars) and we should correctly use the known arg name and not the
string "a" or "b".
2019-12-30 11:21:43 -05:00
Francois Rompre-Lanctot
c0cea013d1 pgraph: Add test for SetValue function 2019-12-18 20:00:47 -05:00
James Shubin
5526bbba64 engine: resources: Add a tftp server and tftp file resource
This adds a tftp server and tftp file resource to help you run a small
pure golang tftp server embedded inside the mgmt resource model.
2019-12-17 03:41:45 -05:00
James Shubin
f0aa96ea8c etcd: Remove the capnslog stuff and switch to zap
Unfortunately, this doesn't give us a way to pass in our own logger
function, and afaict by reading the source, it's not possible because
the necessary methods are private. In any case, this is left as a future
exercise.
2019-12-17 03:40:44 -05:00
James Shubin
e73007c398 etcd: Bump to new 3.4.x version
This moves to the newest etcd release, and also updates the imports to
the new go.etcd.io path. I think this is a bit of a pain, but might as
well get it done.
2019-12-17 02:45:38 -05:00
Jonathan Gold
fdc459ec5b vagrant: Update Vagrantfile
This patch updates the Vagrantfile to Fedora 31, and updates the
install process to match the quick start guide.
2019-11-17 20:25:06 -05:00
Felix Frank
bdb523ece1 lang: funcs: funcgen: Suppress informational messages
Send non-error log messages to stdout rather than stderr. Any messages
outside the main function are expected to be purely informational. By
sending to stdout rather than stderr, they can be discarded during the
build.

Fixes #568
2019-11-17 20:21:52 -05:00
James Shubin
164a9479ad test: Add a new test to the commit message checking
Fix the missing "s" bug.
2019-11-17 20:21:52 -05:00
James Shubin
e18adc781f git: Add a simple gitignore helper
Run `touch 1 && chmod ugo-w 1` to prevent silly scripting bugs from
running without being caught.
2019-11-12 17:33:58 -05:00
Felix Frank
33d89c2739 examples: lib: Update code for urfave/cli v2 2019-11-12 21:07:38 +01:00
Felix Frank
7cc9ab9083 lib: Update for urfave/cli v2 2019-11-12 21:07:38 +01:00
Donald Bakong
4b4b7dc169 engine: resources: Adding tests to file mode 2019-11-08 10:02:30 -05:00
Yohan Belval
71ad5c5f05 examples: lang: Added os check for pkg example fix 2019-11-06 10:20:32 -05:00
Yohan Belval
39368bb5cb examples: lang: Fixed pkg example with cowsay 2019-11-06 10:03:20 -05:00
James Shubin
7a587ee8d1 misc: mkosi: Switch to copy-git-more
I added a new feature in mkosi which got merged in:

31801e89e77188e697ed937ca6b8668fde4c4a4d

This allows us to pull in all of the git repository so that we can see
the version number that comes from git.
2019-11-02 07:34:04 -04:00
James Shubin
77346527f3 docs: Update style guide with more review items
Hopefully this helps new contributors understand review changes and
avoid making them too!
2019-11-01 22:01:38 -04:00
James Shubin
1eba5833d5 engine: resources: Consistency changes and cleanup for file mode
This makes a few consistency changes and cleanups to the file mode
feature so that it's more in style with the rest of the code base.
2019-11-01 22:01:38 -04:00
Derek Buckley
83a747794e engine: resources: Adds symbolic mode to file resource
Adds a symbolic parsing function to the util package for parsing in the
file resource.
2019-11-01 21:57:10 -04:00
Julien Pivotto
3e16d1da46 engine: resources: Add new consul resource
This commit adds a new consul:kv resource which allows us to set and
watch keys inside a consul kv datastore.

This was started by roidelapluie, and was finished during pair
programming with purpleidea.

Signed-off-by: Julien Pivotto <roidelapluie@inuits.eu>
Signed-off-by: James Shubin <james@shubin.ca>
2019-11-01 21:38:08 -04:00
Julien Pivotto
ae1860e859 lang: funcs: Add datetime.format function
Signed-off-by: Julien Pivotto <roidelapluie@inuits.eu>
2019-11-01 19:26:17 -04:00
Julien Pivotto
2ebc8fdf2a lang: funcs: Add datetime.hour function
Signed-off-by: Julien Pivotto <roidelapluie@inuits.eu>
2019-11-01 20:45:45 +01:00
Julien Pivotto
be4023be66 docs: Update resource-guide.md 2019-11-01 10:03:36 -04:00
Julien Pivotto
7f4ad76298 lang: funcs: Fix autogenerated comments
Signed-off-by: Julien Pivotto <roidelapluie@inuits.eu>
2019-11-01 11:18:48 +01:00
Julien Pivotto
0cbfaf98f3 lang: funcs: Support for []byte
Signed-off-by: Julien Pivotto <roidelapluie@inuits.eu>
2019-10-30 22:14:08 +01:00
James Shubin
631124e658 lang: funcs: Add nitpicks from funcgen
Discussed nitpicks with roidelapluie to clean up slightly for
consistency.
2019-10-30 08:51:11 -04:00
Julien Pivotto
1685ee1ecb lang: funcs: Autogenerated a lot of new functions
Signed-off-by: Julien Pivotto <roidelapluie@inuits.eu>
2019-10-30 08:42:53 -04:00
James Shubin
9b4d11f220 lang: funcs: Move convert into correct folder
This got merged in the wrong folder by accident.
2019-10-30 06:21:34 -04:00
James Shubin
46a71296a9 engine: resources: Add a Purge option to the file resource
This adds a "purge" parameter to the file resource. To do this, we have
to add the API hooks so the file resource can query other resources in
the graph to know if they are present, and as a result whether they
should be excluded from the purge or not.

This is useful for when we have a managed directory with some managed
contents. If a managed file is removed from the directory, then it will
be removed by the file (directory) resource if it has Purge set.
Alternatively, you can use the Reverse meta param, which is sometimes
preferable for this use case and sometimes not. This will be discussed
elsewhere.

This also adds a bunch of tests for this feature.

This also makes a few somewhat related cleanups in the file code.
2019-10-29 11:54:08 -04:00
James Shubin
1285588b62 engine: resources: Fix file absent helper
We should check this nil case too.
2019-10-29 07:15:43 -04:00
James Shubin
d96392f65e engine: resources: Improve error message in test
Nothing major, just improve testing here.
2019-10-29 07:15:43 -04:00
James Shubin
d1c5a736ae engine: resources: Allow nil helper functions in tests
This reduces some of the boilerplate when writing resource tests.
2019-10-29 07:15:43 -04:00
James Shubin
6b1e038c5c engine: resources: Add file exists helper
This will allow us to test even more things!
2019-10-29 07:15:43 -04:00
James Shubin
eaab1aae28 engine: graph, resources: Add filtered graph function
This lets a resource query the resource graph in a controlled way.
2019-10-29 07:15:43 -04:00
James Shubin
31030343a2 engine: Add graph queryable trait
This is a trait that specifies that your resource can be queried by
others. You can either enable it plainly to add wholesale access by
everyone, or you can implement your own Allow method to limit what is
permitted.
2019-10-29 07:15:43 -04:00
James Shubin
325ca03a13 engine: graph: Pass through the graph struct
We want to use it in the resources.
2019-10-29 07:15:43 -04:00
James Shubin
dea8e63df2 util: Add more tests to HasPathPrefix
We need to ensure this behaviour in a future commit, so might as well
double check that this works as expected!
2019-10-29 07:15:43 -04:00
James Shubin
58421fd31a engine: resources: Add fragments support to file resource
This adds a "fragments" mode to the file resource. In addition to
"content" and "source", you can now alternatively build a file from the
file fragments of other files.
2019-10-29 07:15:43 -04:00
James Shubin
b961c96862 lang: Include automatic edges in our test case
When running this test, we didn't attempt to build any automatic edges.
Since we'd like to test this here as well, let's add it.
2019-10-24 04:30:49 -04:00
Donald Bakong
2d23c1b0f3 lang: funcs: Add to_int and to_float functions 2019-10-22 08:34:29 -04:00
James Shubin
06952c224b engine: resources: nspawn: Use populate variable
We referred to the wrong variable. Not a major bug, but would produce a
useless or confusing error message otherwise.
2019-10-21 07:38:42 -04:00
Donald Bakong
2ea492c965 docs: Fix error on language-guide.md 2019-10-19 18:44:56 -04:00
Donald Bakong
dbf84f6879 docs: Fix typo on language-guide.md 2019-10-19 18:44:11 -04:00
James Shubin
0fa3d6c462 github: Update funding information file
Just got added to the GH sponsors thing and there are more fields
available in this file now. Let's hope this works!
2019-10-09 19:56:19 -04:00
James Shubin
d57f7aa03f misc: Specific mkosi fixes for centos-7
Seems we need golang from epel, to mask out the old git version, and to
workaround mkosi bugs.
2019-10-05 01:49:42 -04:00
Jimmy Tang
d64f9f5401 misc: Add CentOS 7 rpm build 2019-10-04 23:47:12 -04:00
James Shubin
a3029afc41 art: Rename art file to clarify it's about Winnie the Pooh
Pretty cool that we have our first meme =D Have a look in the FAQ to see
the real reason we watch our resources. (TL;DR: It allows us to make
graph changes very fast.)
2019-10-04 08:01:23 -04:00
James Shubin
6a7d904fae misc: Improve tagging script
This way we can push the tag *after* all the builds succeed. If
something goes wrong, we can always delete our local tag and try again.
2019-10-04 06:49:04 -04:00
James Shubin
d4043d3f86 misc, make: Add full file path into fpm script
This is needed for our fancier, unique file names.
2019-10-04 06:43:17 -04:00
James Shubin
b4902a4f58 make: Add a unique token to the package file name
This unique token is necessary so that storing the files in the same dir
(basically a GitHub release) or in the SHA1SUMS file doesn't cause a
conflict.
2019-10-04 06:06:44 -04:00
James Shubin
ffe402f201 misc: Add fedora-30 mkosi+fpm build environment
Good example of how to add a new distro or version.
2019-10-04 06:02:08 -04:00
James Shubin
09cc7da282 misc: Add proper archlinux prefix in build script 2019-10-04 06:01:23 -04:00
James Shubin
2d2dad41f4 todo: Update the TODO file so that it has a sane purpose
We stored some stuff in GitHub, and some stuff here. We can keep using
this, but let's do it for the stuff that hasn't changed in a while.
2019-10-04 04:11:26 -04:00
bjanssens
5f7c0a86dd art: Add the requested art
Signed-off-by: bjanssens <bjanssens@inuits.eu>
2019-10-04 09:23:20 +02:00
Donald Bakong
fc1c631c98 engine: resources: Change Res API from Compare to Cmp
This will be done by refactoring the current method, to return an error
message instead of a boolean value. This will also update a typo on the
user res.
2019-09-27 18:10:58 -04:00
James Shubin
89bdafacb8 misc: Refactor Makefile slightly
We could make this even better in the future with lists.
2019-09-23 06:57:26 -04:00
James Shubin
73b6b3f129 misc: Remove old image building cruft 2019-09-23 06:50:26 -04:00
James Shubin
b2a495f593 misc: Add mkosi target for ubuntu bionic
The name of these is pretty weird.
2019-09-23 06:50:00 -04:00
James Shubin
65ee904377 misc: Work around old golang in ubuntu
Hopefully this helps.
2019-09-23 06:48:55 -04:00
James Shubin
13f59230b5 misc: Split Makefile PHONY target into multiple lines
AIUI this is valid make. Please correct me if I'm wrong.
2019-09-23 06:48:55 -04:00
James Shubin
36d2a0de1e misc: Make mkosi building suitable for different distro versions
We'd like to be able to build both Fedora N and N-1 at the same time if
possible. This makes it more generally applicable for this scenario, as
well as for other distros.
2019-09-23 06:48:55 -04:00
James Shubin
a4db9fc8e5 misc: Add mkosi based package building with fpm
Building distro packages is great, however if they aren't built in the
correct environment with the associated dependencies, then they won't
work properly on those distros.

This patch adds an `mkosi` based image building environment that builds
the packages in their respective distros, and then copies them out into
our releases directory.

You'll now want to `make tag && make mkosi && make release` to get a new
release out. We use a small hack to trick the `make release` portion to
not re-build the distro packages if they're already present in the
releases/ directory for that version.

This commit depends on a very recent version of mkosi (it was tested
with git master) and also depends on two currently unmerged patches:
https://github.com/systemd/mkosi/pull/363 and
https://github.com/systemd/mkosi/pull/365
2019-09-20 12:32:41 -04:00
James Shubin
9dae5ef83b engine: resources: Improve the file res and add strict state
This might be slightly controversial, in that you must specify the state
if a file would need to be created to perform the action. We no longer
implicitly assume that just specifying content is enough. As it turns
out, I believe this is safer and more correct. The code to implement
this turns out to be much more logical and simplified, and does this
removes an ambiguous corner case from the reversed resource code.

Some discussion in: https://github.com/purpleidea/mgmt/issues/540

This patch also does a bit of related cleanup.
2019-09-14 16:07:53 -04:00
James Shubin
e8842a740c lang: Remove duplicate log message
Looks like we had two copies of the same message by accident.
2019-09-11 04:26:15 -04:00
James Shubin
0d3807ad09 lang, test: Fix copy paste error with log message
This changes this to the correct error message.
2019-09-11 04:26:15 -04:00
James Shubin
5c27a249b7 engine: resources: Add reversible API and file resource
This adds the first reversible resource (file) and the necessary engine
API hooks to make it all work. This allows a special "reversed" resource
to be added to the subsequent graph in the stream when an earlier
version "disappears". This disappearance can happen if it was previously
in an if statement that then becomes false.

It might be wise to combine the use of this meta parameter with the use
of the `realize` meta parameter to ensure that your reversed resource
actually runs at least once, if there's a chance that it might be gone
for a while.

This patch also adds a new test harness for testing resources. It
doesn't test the "live" aspect of resources, as it doesn't run Watch,
but it was designed to ensure CheckApply works as intended, and it runs
very quickly with a simplified timeline of happenings.
2019-09-11 03:40:22 -04:00
James Shubin
7e41860b28 docs: Add missing docs on the rewatch and realize meta params
Sometimes it's hard to keep this in sync.
2019-09-11 03:40:22 -04:00
James Shubin
43ff92bbe7 engine: resources: Clean up test log message 2019-09-11 03:40:22 -04:00
James Shubin
28adc7e563 engine: resource: Refactor helper functions
Maybe we can use them in other tests too.
2019-09-11 03:40:22 -04:00
James Shubin
9788411995 engine: resources: Add another validation check
This simple check should prevent some silly mistakes and make the logic
easier for other parts of the code that won't have to worry about this
pattern.
2019-09-11 03:40:22 -04:00
James Shubin
0c9e8cc50e engine: resources: Change the default file state
The default file state should be undefined. This is important because if
a reverse scenario that doesn't specify the state gets given this
default, it will be as if it was specified explicitly, which wouldn't
necessarily be what we want. Instead, an undefined state should
implicitly cause a file to get created if there's a reason to do so,
such as if content or another attribute is specified.

Hopefully this change doesn't introduce any bugs in the CheckApply code,
if it does, then it was due to a lack of implicit file creation.
2019-09-11 03:16:57 -04:00
James Shubin
34d572c523 engine: Improve the way we make a unique res path token
This is needed in the state directory.
2019-09-11 03:16:57 -04:00
James Shubin
011b496b3f engine: resources: Ensure the Kind and Name methods work
Triple check these work after decoding, by adding a test.
2019-09-11 03:16:57 -04:00
James Shubin
12b906eac6 engine: Refactor state dir into a separate function
This lets us re-use it, and know the path is fixed.
2019-09-11 03:16:57 -04:00
James Shubin
20937d05c3 engine: resources: file: Add undefined file state and validate it
We should consider using *string instead of the empty string, but let's
keep the diff smaller for now.
2019-09-11 03:16:57 -04:00
James Shubin
4943d37ccf engine: resources: file: Use constants for state values
More robustness is yay!
2019-09-11 03:16:57 -04:00
James Shubin
3a8fd215de engine: resources: file: Add Copy method to file res
This lets us implement the CopyableRes interface.
2019-09-11 03:16:57 -04:00
James Shubin
87572e8922 test: Catch capitalized error messages in tests 2019-09-06 03:28:49 -04:00
James Shubin
f1eedc7a01 lang: Clarify error message about missing field
User probably just mistyped a field name. Make that clear.
2019-09-06 03:28:49 -04:00
Donald Bakong
b79e48dd77 docs: Fix typo on quick-start-guide.md 2019-08-25 22:53:43 -04:00
James Shubin
18872194af misc: Warn users with weird computers
A user seemed to experience a weird golang issue when they had deps from
both package managers installed. I won't block or fail their install,
but we can print a warning message so that someone sees it in their
logs.
2019-08-23 22:10:34 -04:00
James Shubin
bafd7ba282 misc: Use apt install of apt-get where possible
The future is now!
2019-08-23 22:10:12 -04:00
Donald Bakong
b186481181 pgraph: Add a test for FindEdge() function 2019-08-08 00:43:26 -04:00
James Shubin
09ca6d11ad lang: funcs: Module name should be public
For consistency with the rest of the core functions.
2019-07-29 11:17:43 -04:00
James Shubin
e68e4e786d docs: Add newly recorded talks and blog post 2019-07-26 06:52:01 -04:00
James Shubin
ee638254c3 lang: Remove the specialized info structs
Since this was an early form of the modern data struct, remove those and
pass in the correct data. This is also important in case we have
something more complex inside our string interpolation!
2019-07-26 04:20:04 -04:00
James Shubin
1e678905c4 util: Fix typo 2019-07-26 04:20:04 -04:00
James Shubin
10804c4b25 lang: Improve the gapi copying
We hit a weird bug where dirs would not get copied properly. I thought
the solution might be to add the missing dirs so they'd get a proper
mkdir, but in the end that didn't work well, so we just use `mkdirall`
and that seems to work. Let's leave it like this for now. Some of the
previous work for that is in the previous commit.
2019-07-26 04:20:04 -04:00
James Shubin
4bf9b4d41b util: Add some path helper functions
In the end, I'm not sure how useful these will be, but we'll keep them
in for now.
2019-07-26 04:20:04 -04:00
James Shubin
1161872324 etcd: fs: Errors should start with lower case 2019-07-26 04:20:04 -04:00
James Shubin
98cb570896 util: Add new mkdirall variants for the copy functions
This adds versions that recursively `mkdir` and all don't error as
easily. This works around some bugs we were having with file copying.
2019-07-26 04:20:04 -04:00
James Shubin
ed4ee3b58e lang: funcs: Add deploy package with readfile related functions
This adds a readfile function to actually access files from our deploy.
A fun side effect is that we can even access our own code! In general,
it's a good reminder that you should only run trusted code on your own
infrastructure. This also includes a fancy new test case.
2019-07-26 03:38:26 -04:00
James Shubin
066048f4de lang: Pass through the Fs and the FsURI
This should give us options as to how a function should interact with an
FS. I feel like it's cleaner to go through the World API, and passing in
the FsURI lets us do that, but I passed in the Fs at the same time in
case it's useful for some reason. I think using it is a boundary
violation, but it's just a hunch. Does anything break when we move from
one deploy to the next?
2019-07-26 03:07:08 -04:00
James Shubin
4b6b91c08b lang: Make sure to call Init for functions that arrive via import
We weren't calling Init on some functions which should have had this
done. I'm not sure whether this is the right place, or if it should be
elsewhere as part of the scope building process. Good enough for now.
2019-07-22 06:49:02 -04:00
James Shubin
2980523a5b lang: Add a new function interface to accept data
Sometimes certain internal functions might want to get some data from
the AST or from something relating to the state of the language. This
adds a method to pass in that data. For now it's a very simple method,
but we could generalize it in the future if it becomes more useful.
2019-07-22 06:46:04 -04:00
James Shubin
f2f9c043bf lang, gapi: Work around a copy bug in the deploy
It seems when we had a files/ dir that we added to our deploy, it would
get copied into /files/files/whatever instead of /files/whatever where
it should be. Hopefully this works around the issue forever.
2019-07-22 06:40:47 -04:00
James Shubin
5d59cfd2c9 util: Ensure the afero copy function is working as intended
The destination should be a dir sometimes.
2019-07-22 06:38:02 -04:00
James Shubin
f94474e24f lang: Add the world implementation to our test suite
This allows our tests to actual run the World API in them.
2019-07-22 06:36:37 -04:00
James Shubin
a63fc6d9ba util: Add a remove path suffix util function
This pairs with a similar one we already had.
2019-07-22 06:35:13 -04:00
James Shubin
076adeef80 lang: funcs: Fix a copypasta error with the not equals operator
Woops, sorry!
2019-07-22 06:08:37 -04:00
James Shubin
a0e756317c lang: Add tests for slow unification
These used to be cases where our algorithm was unusably slow.

Thanks to foxxx0 for the report!
2019-07-21 03:15:06 -04:00
James Shubin
252cb5f2f3 lang: Detect windows style CR and return a better error
If you get a sneaky \r in your code, the error just looks like
whitespace, so this way we can warn you explicitly.
2019-07-21 03:10:21 -04:00
James Shubin
64288b4914 lang, test: Inline some overly indented tests
Sometimes you're busy hacking and it's nice for future you to fix up
your code!
2019-07-21 01:19:15 -04:00
James Shubin
9ca6c6a315 test: Split up long tests into multiple sub tests again
I think we need this for non --race tests too.
2019-07-21 00:55:36 -04:00
James Shubin
3651ab5c0c lang: Add more tests for function 2019-07-20 22:27:21 -04:00
James Shubin
b3f15e1ddc lang: Add more tests for class and include 2019-07-20 01:33:42 -04:00
James Shubin
da2a5f72bd lib: Update dep for uuid
Apparently the package has moved.
2019-07-17 04:07:24 -04:00
James Shubin
591e6b68e0 test: Split up long tests into multiple sub tests
Hopefully this avoids the timeouts running the lang package.
2019-07-17 02:45:04 -04:00
James Shubin
0119abdcdd travis: Try to work around CI slowdowns
I think they throttle strangely on their garbage machines.
2019-07-17 01:21:29 -04:00
James Shubin
e57ca15330 lang: Avoid running graphviz in tests by default
This will help travis actually run the tests faster.
2019-07-17 01:21:29 -04:00
James Shubin
f53376cea1 lang: Add function values and lambdas
This adds a giant missing piece of the language: proper function values!
It is lovely to now understand why early programming language designers
didn't implement these, but a joy to now reap the benefits of them. In
adding these, many other changes had to be made to get them to "fit"
correctly. This improved the code and fixed a number of bugs.
Unfortunately this touched many areas of the code, and since I was
learning how to do all of this for the first time, I've squashed most of
my work into a single commit. Some more information:

* This adds over 70 new tests to verify the new functionality.

* Functions, global variables, and classes can all be implemented
natively in mcl and built into core packages.

* A new compiler step called "Ordering" was added. It is called by the
SetScope step, and determines statement ordering and shadowing
precedence formally. It helped remove at least one bug and provided the
additional analysis required to properly capture variables when
implementing function generators and closures.

* The type unification code was improved to handle the new cases.

* Light copying of Node's allowed our function graphs to be more optimal
and share common vertices and edges. For example, if two different
closures capture a variable $x, they'll both use the same copy when
running the function, since the compiler can prove if they're identical.

* Some areas still need improvements, but this is ready for mainstream
testing and use!
2019-07-17 00:27:09 -04:00
James Shubin
4f1c463bdd misc: Add graphviz deps for travis
Some tests run graphviz.
2019-07-17 00:27:09 -04:00
James Shubin
6643a3d937 lang: Add function types to the yacc type parser
Hopefully our type unification algorithm will be sufficiently good that
you never need to actually specify the function type, but it's useful
for testing and completeness.
2019-07-12 16:46:08 -04:00
James Shubin
da8cb40242 lang: If the test fails earlier than expected, exit early
If a test failed in stage 2 (fail2) instead of an expected fail in stage
3 (fail3) then it would continue running, which was an undefined
behaviour in our API. IOW we should not run Unify if SetScope failed.
This patch adds these additional checks to ensure our tests are more
robust.
2019-07-12 16:46:08 -04:00
James Shubin
4c6d304e60 lang: types: Improve ComplexCmp function
This improves the ComplexCmp function so that it can compare partial
types to variant types. As a result of this improvement, it actually
ended up simplifying the code significantly. This also added a test
suite for this function. This fix was important for tricky type
unification problems.
2019-07-12 16:46:08 -04:00
James Shubin
99d3ef42e9 lang: Name the expr call graph differently
It was wrongly named func instead of call, although this doesn't
actually matter in terms of code execution.
2019-07-12 16:46:08 -04:00
James Shubin
e2289dc2a0 lang: funcs: Add better logging to the function engine 2019-07-12 16:46:08 -04:00
James Shubin
9b4f50cde9 lang: Add the NamedArgs interface
This lets you specify which args are being used in the general function
API, which can make code readability and debugability slightly better.

In an ideal world, we wouldn't need this at all, but I can't figure out
how to avoid it at the moment, so we'll include it for now, as it's
always easy to delete if we find a more elegant solution.
2019-07-12 16:46:08 -04:00
James Shubin
fe64bd9dbb lang: Move type duplicates checker into a separate package 2019-07-12 16:46:08 -04:00
James Shubin
0991264c8c lang: funcs: Use the correct arg names when running a pure func
We were using the default argnames when the actual list of names was
available. Use these instead, and validate that we have the correct
number of them.
2019-07-12 16:46:08 -04:00
James Shubin
3b608ad544 lang: funcs: Verify that simple polyfunc list is not empty
This could prevent some incorrect usage of the simplepoly API.
2019-07-12 16:46:08 -04:00
James Shubin
3f1a379908 lang: Use the short flag to list subtests
If you want to know which test to run, it's not always obvious, so by
adding the -short flag, we'll get a listing that we can use! You'll need
to add -v as well so that the output actually displays.
2019-07-12 16:46:08 -04:00
Adam Sigal
61a67dae29 pgraph: Add a test for FindEdge() and HasVertex() functions
Added a new test called TestFindEdge1(). In it, a 7-node
graph is created on which to test the aforementioned functions.
2019-07-11 22:05:13 -04:00
John Hooks
609aefd808 lib: Support for systemd STATE_DIRECTORY or XDG cache dir
If running mgmt from a systemd unit, this enables the
STATE_DIRECTORY environment variable to be used for creating the
cache directory defined by StateDirectory= in the unit file. It
also enables the XDG_CACHE_HOME environment variable to be used.
If the user isn't root and the environment variable isn't set,
it will use the default XDG_CACHE_HOME directory.
2019-07-08 21:19:33 -04:00
Felix Frank
191a2495a5 engine: resources: mount: Fix the dbus call for reloading systemd
The Reload method cannot just be invoked on the administrative DBus
object. Just like the method for reloading specific units, it needs
to be invoked on the proper DBus service, addressing the proper object
and using the right interface.

Added an additional constant for the systemd DBus service. Even though
it shares the same value as the interface base name, this is
happenstance and it's technically incorrect to open a connection to an
interface name. The connection needs a service name.

Fixes #509
2019-06-04 23:44:57 +02:00
James Shubin
a235b760dc docs: Fix typo and grammar issue 2019-05-20 09:53:19 -04:00
James Shubin
e4eb3c23a2 lang: funcs: core: Allow nested system imports
We were passing the wrong module name for system imports. This is now
fixed, includes an example, and some tests!
2019-05-20 09:23:28 -04:00
James Shubin
12582e963d lang: funcs: core: Make module names public
This is needed for when we have nested modules.
2019-05-20 08:45:43 -04:00
James Shubin
d5074871c7 examples: lang: Add a unicode example 2019-05-15 04:13:20 -04:00
James Shubin
e0d024ac95 examples: lang: Update autoedges example
The /dev/null thing isn't needed anymore. Also make it easier to change
noop value.
2019-05-15 03:51:01 -04:00
James Shubin
7a756cacb9 engine: graph: Add a mutex around waits map access
If you ran some extremely absurd code, it turns out you can cause a
race. This was found by roiedelapluie experimenting! In this case, it
would panic with: fatal error: concurrent map read and map write. This
patch adds the mutex to avoid this rare race.
2019-05-14 10:53:36 -04:00
Jan Martens
3c1da423fa engine: resources: nspawn: Trim possible systemd version suffix 2019-05-14 16:00:26 +02:00
James Shubin
38dfaa1caa docs: Update FAQ to mention go mod 2019-05-14 06:18:53 -04:00
James Shubin
a050cff50f docs: Add build issue to FAQ
Some new users might experience this if they setup their $GOPATH
incorrectly.
2019-05-13 07:10:36 -04:00
James Shubin
93c1b37aab lang: funcs: Add a mod function to the math package
This should make flip-flop functions easy to write.
2019-05-13 06:30:15 -04:00
James Shubin
01d4226c4a docs, readme: Improve new user experience
This hopefully improves some docs for new users, and makes releases more
easily available.
2019-05-06 07:56:38 -04:00
James Shubin
fc6032d3b7 lang: funcs: Add a weekday function to the datetime package
This returns the day of the week. It also includes a helpful example
demonstrating how this functionality can be fun!
2019-05-06 06:59:50 -04:00
James Shubin
43839d1090 all: Switch the --lang syntax to use argv instead
It was a bit awkward using `mgmt run lang --lang <input>` so instead, we
now drop the --lang, and read the positional argv for the input.

This also does the same for the --yaml frontend.
2019-05-05 11:10:47 -04:00
James Shubin
b3632584c3 etcd: Move to etcd v3.3.13
This makes a small jump to the new etcd stable release. This isn't a
major difference, but it includes an important patch in
7814718c73149e2bbb9517cd02edb8332b621d86 which caused mgmt users to
scratch their heads, since it wasn't obvious that etcd was doing a Fatal
instead of a Panic or an error.
2019-05-05 09:32:04 -04:00
James Shubin
e9257580cd misc: Update to golang 1.11.x
Bump to the newer version.
2019-05-05 09:32:04 -04:00
James Shubin
e3cc6309ea lib: Fix gofmt regression in golang 1.11.x
Golang has many exceptions to its "compatibility promise", including the
gofmt output. The fact that they change it arbitrarily for things like
this is absurd. (Remove the patch and run `gofmt` to see for yourself.)

This change re-worked the comment, since include the `gofmt` suggested
line break makes absolutely no sense, and is not convenient.
2019-05-05 09:32:04 -04:00
James Shubin
17fd625f7f lang: types: Workaround stringer regression in golang 1.11
Here's a fix for another golang regression in 1.11.x which wasn't needed
before! More info in: https://github.com/golang/go/issues/31843
2019-05-05 09:32:04 -04:00
James Shubin
d1ecfd8657 test: Fix typos, these aren't cats
Miau!
2019-05-05 09:32:04 -04:00
James Shubin
4aa3cfad40 lang: Add var prefix to var expr to avoid ambiguity 2019-05-05 09:32:04 -04:00
James Shubin
3bcb697662 lang: funcs: structs: Make error message more precise
This should prevent ambiguity with other similar or identical error
messages.
2019-05-05 09:32:04 -04:00
James Shubin
88318b73e4 lang: Print the actual stream error on test failure
This is useful for debugging.
2019-05-05 09:32:03 -04:00
James Shubin
2f7e202f40 lang, lang: types: Add automatic stringer generation
It's more useful if we know the string representation of Kind's.
2019-05-05 09:32:03 -04:00
James Shubin
310239e707 lang: types: Remove unnecessary prefix in generated kinds
This will make displayed errors cleaner.
2019-05-05 09:30:13 -04:00
James Shubin
4de75373dd pgraph: Use pointers for unique vertex identifiers
This will build more accurate graphs, since we could have duplicated
vertex names for distinct vertices. This now builds the correct
topology, even if the labels are duplicated.
2019-04-29 16:08:36 -04:00
James Shubin
c0d329e6d8 pgraph: Quote graphviz strings properly
If strings include quotes, this previously didn't work.
2019-04-29 16:08:36 -04:00
Johan Bloemberg
8a0840d35b lang: funcs: Add uptime implementation for macOS 2019-04-29 16:07:21 -04:00
Ward Vandewege
f9bb9ef33e docs: Fix link to puppet guide 2019-04-29 06:11:13 -04:00
Christian Rebischke
acb2a5d2b0 lang: funcs: Add ArchLinux family detection
Signed-off-by: Christian Rebischke <chris@nullday.de>
2019-04-25 17:05:25 +02:00
James Shubin
63ef11c708 lang: Add a new type unification test
I wanted to make sure that the type unification algorithm restricts the
implementation of the class when included, when one of the polymorphic
types is specified with a fixed type. It seems this works! I had the
idea for this test while walking around aimlessly.
2019-04-24 03:46:21 -04:00
James Shubin
d70bbfb5d0 lang: unification: Improve type unification algorithm
The simple type unification algorithm suffered from some serious
performance and memory problems when used with certain code bases. This
adds some crucial optimizations that improve performance drastically.
2019-04-23 21:21:42 -04:00
James Shubin
97d60ac98d lang: Quote printed strings
This quotes printed strings that contain special characters such as
newline. This changes the output of some tests, but makes future tests
that include a raw \n more appropriate.
2019-04-23 21:03:02 -04:00
Jonathan Gold
8f1f5d33fd engine: resources: mount: Restart remote-fs target 2019-04-23 16:24:49 -04:00
Wouter Dullaert
d65c85c19f cli: Removed obsolete no-watch-config flag
Having it around creates the expectation that by default mgmt will put a watch
on the config.
2019-04-22 13:42:27 +02:00
James Shubin
22d893fc1e test: shell: Increase etcd timeouts for slow travis again
Increase this one too...
2019-04-21 20:11:38 -04:00
James Shubin
806d2f6a4a lang: Fix import scoping issue with classes
When include-ing a class, we propagated the scope of the include into
the class instead of using the correct scope that existed when the class
was defined and instead propagating only the include arguments in.

This patch fixes the issue and adds a ton of tests as well. It also
propagates the scope into the include args, in case that is needed, and
adds a test for that as well.

Thanks to Nicolas Charles for the initial bug report.
2019-04-21 19:49:38 -04:00
James Shubin
fc3baa28d6 lang: funcs: Add regexp package and match function
This adds a simple regexp match function. This will be useful for
regexp based name classification, if you're into that sort of thing.
2019-04-16 21:42:32 -04:00
James Shubin
eba45e6207 lib, gapi: Display deploy ID to add some clarity
This should make it easier to understand exactly when a new deploy
starts.
2019-04-16 18:11:32 -04:00
James Shubin
272fd3edc3 test: shell: Increase etcd timeouts for slow travis
We need a real test environment that's not travis.
2019-04-16 18:08:48 -04:00
James Shubin
5ad8b33aa7 etcd: Make error more specific
This should clarify which member remove branch we're in if we error.
2019-04-16 18:08:39 -04:00
James Shubin
cacd14fcf8 util: Make test more resistant to races
This doesn't guarantee which print statement runs first, so the last
part of it can race. Adding a sleep makes this highly unlikely.
2019-04-11 21:43:48 -04:00
James Shubin
859e4749ae lib: Clean up logging
Since most of our logging goes through a single Logf command, we don't
need the file name information any more. Our hierarchial logging is
sufficient enough.

Eventually we will replace the top-level logger with a more visually
capable logging fixture.
2019-04-11 21:43:48 -04:00
James Shubin
a5842a41b2 etcd: Rewrite embed etcd implementation
This is a giant cleanup of the etcd code. The earlier version was
written when I was less experienced with golang.

This is still not perfect, and does contain some races, but at least
it's a decent base to start from. The automatic elastic clustering
should be considered an experimental feature. If you need a more
battle-tested cluster, then you should manage etcd manually and point
mgmt at your existing cluster.
2019-04-11 21:43:48 -04:00
James Shubin
fb275d9537 test: Skip net test in travis
Travis fails intermittently, and I have no idea what's wrong with their
infra or what's using this address, so skip it here.
2019-04-10 22:55:02 -04:00
James Shubin
88f7b7e786 docs: Add faq entry about locked binary
Doesn't happen very often, but add it in case someone is curious or uses
search engine foo to figure out the answer.
2019-04-05 17:50:25 -04:00
James Shubin
30402effa9 util: Add context utility functions
This adds a utility function to close a context via a closed signalling
channel, and also functions to wrap and unwrap a wait group into and out
of a context.
2019-04-02 15:30:34 -04:00
James Shubin
7d96623f06 util: Add safe easy ack that allows multiple ack's
Just another sync utility to make code more readable.
2019-04-02 07:07:36 -04:00
James Shubin
398706246e util: Add subscribed signal primitive
Add a little sync primitive to our utility library. This should
hopefully make some of the future code easier to deal with.
2019-04-02 07:07:36 -04:00
James Shubin
6628fc02f2 util: Add context wait signal to easy exit
Add an alternate way to wait for a signal. This just makes code look a
bit cleaner and less cluttered.
2019-04-02 07:07:36 -04:00
Michael Schubert
e2fa7f59a1 docs: Fix link 2019-04-02 10:33:41 +02:00
James Shubin
d5b7dc0acc github: Add funding information 2019-03-24 15:37:54 -04:00
James Shubin
e4d874cc69 engine: resources: Remove named return params
Named return params aren't a favourite feature of mine, and they're
rarely used in the resource writing. They keep popping up because I once
used them and we've been copying and pasting code ever since. Get rid of
them all to help prevent the unnecessary spread.
2019-03-24 15:30:02 -04:00
James Shubin
80a0abeead docs: Add FAQ entry about root requirements 2019-03-24 15:16:14 -04:00
James Shubin
0df2d46ca7 lib: Add static hello message 2019-03-24 15:11:01 -04:00
James Shubin
07f542b4d7 legal: Happy 2019 everyone...
Done with:

ack '2018+' -l | xargs sed -i -e 's/2018+/2019+/g'

Checked manually with:

git add -p

Hello to future James from 2020, and Happy Hacking!
2019-03-24 15:08:50 -04:00
Mitch Fossen
7db3e8556a lang, funcs: Remove deprecated syscall import 2019-03-24 14:46:58 -04:00
Felix Frank
dc03e67b81 docs: Slightly clarify parameter defaults 2019-03-24 14:44:51 -04:00
Adam Sigal
e587324b81 faq: Amended faq mailing list information
There was outdated information concerning the mailing list. I amended
this and added a link.
2019-03-19 14:39:13 -04:00
James Shubin
65a66492f4 docs: Move faq entry to more appropriate resource docs 2019-03-17 17:54:48 -04:00
James Shubin
17602d7065 docs: Add two faq entries 2019-03-17 17:54:48 -04:00
James Shubin
ae56261961 engine: util, resources: virt: Clean up virt resource
Do some cleanups which were long overdue.
2019-03-15 15:24:40 -04:00
James Shubin
c4f57608d0 test: Port yaml test to mcl 2019-03-15 13:01:50 -04:00
James Shubin
753d1104ef util: Port all multierr code to new errwrap package
This cleans things up and simplifies a lot of the code. Also it's easier
to just import one error package when needed.
2019-03-12 16:51:37 -04:00
James Shubin
880652f5d4 util: Port all code to new errwrap package
This should keep things more uniform.
2019-03-12 16:49:01 -04:00
James Shubin
54c81d6bb2 engine, pgp: Fixup incorrect error usage
Small fixups found by the next commit.
2019-03-12 16:49:01 -04:00
James Shubin
2bf43eae24 test: Improve test depth
Make sure we're catching everything, including new, deeper code.
2019-03-12 16:49:01 -04:00
James Shubin
58961d23bb test: Improve test depth
Make sure we're catching everything, including new, deeper code.
2019-03-12 15:50:30 -04:00
James Shubin
6044ade373 util: Add errwrap package
Simplify working with errors across our code base. Instead of constantly
importing the necessary error helpers, assemble them all into one
package and import and use that as needed.
2019-03-12 15:45:39 -04:00
James Shubin
da1c96c6fd examples: lang: Refresh examples
Removed two old examples which were no longer valid.
2019-03-09 18:54:50 -05:00
James Shubin
5bbb474db6 engine: resources: Clean up KV resource 2019-03-09 18:48:26 -05:00
James Shubin
a0c909914d lang: funcs: Don't allow interpolation in printf format string
We'd like to pre-compute the interpolation if we can, so that we can run
this code properly... For now, we can't so it's a compile time error...
Hopefully we can remove this restriction in the future. The problem is
the string must be a constant, or it would be possible to switch it from
"%d %s" to "%s %d %d" or anything that changes the type signature.
2019-03-09 18:06:18 -05:00
James Shubin
170e56b34a lang: Improve test case with more specific errors 2019-03-09 17:37:58 -05:00
James Shubin
de43569fa2 engine, lang: Improve send/recv significantly
Part of this was rotten, and not fully functional. This fixes the rot,
adds some tests, and improves the type checking that occurs when sending
and receiving values. In addition, a significant portion of this happens
at compile time.

There is still more work to be done here, but this should get us a good
chunk of the way for now.
2019-03-09 17:37:58 -05:00
James Shubin
aa6b701b77 lang: Improve the test case infra so it can detect different errors
Let the tests detect which specific error we want to fail on.
2019-03-09 17:14:45 -05:00
James Shubin
d69eb27557 lang: Small fixes about send/recv 2019-03-09 16:07:22 -05:00
James Shubin
0ca57d6a09 examples: lang: Add a basic file example 2019-03-09 16:05:12 -05:00
James Shubin
4c104d55cb engine: util: Add a new utility function for send/recv
This new utility function makes verifying send/recv struct comparisons
consistent. Unfortunately it doesn't yet support coercing from *string
to string or from string to *string.
2019-03-09 16:03:30 -05:00
James Shubin
8a8215fabe engine: util: Improve StructTagToFieldName and add tests
This improves this function to make it more generic.
2019-03-09 16:02:33 -05:00
James Shubin
4badeafb98 engine: resources: Add missing struct tags
These are needed for send/recv.
2019-03-09 16:01:25 -05:00
James Shubin
7cb79bec49 engine: resources: Replace the cached values with a live calculation
This replaces the static obj.path and obj.isDir with live variants. I
don't know why I ever cared about caching these before, and if we ever
care we can memoize properly in the future.

This caused a small bug to actually be masked in the gob code. It is now
fixed in the previous commit.
2019-03-06 10:08:14 -05:00
James Shubin
8da0da02d9 engine: traits: Make encoded fields public
The fields that we might want to store with encoding/gob or any other
encoding package, need to be public. We currently don't use any of these
at the moment, but we might in the future.
2019-03-06 10:08:14 -05:00
James Shubin
efef260764 engine: resources: Improve file Cmp function 2019-03-06 07:09:39 -05:00
James Shubin
a56991d081 engine: resources: Remove possible panic from within file res
Not sure how I let this in, but we should never do this. Hopefully the
Validate should catch this issue in advance, and if not, at least we'll
only error.
2019-03-06 07:09:39 -05:00
James Shubin
f0196540ab resources: file: Make some small cleanups to file res
This does some small cleaning for consistency, since I haven't reviewed
this code in a long while.
2019-03-06 07:09:39 -05:00
James Shubin
426b15313e engine: resources: Fix missing file when specified without content
If the file res was defined with state => "exists" but no content
specified, it was not created. This patch fixes this bug and adds a test
and an example.
2019-03-06 07:09:39 -05:00
James Shubin
11fc55d679 lang: funcs: Add a new test for readfile and fix a small bug
This adds a new test for readfile. Interestingly, it actually caught a
small bug, which was also fixed with this commit. I think the bug was
actually always masked, because it only occurred on shutdown, and in
this case we often don't care about how the stream exited, but it's a
good example of how a test case focused on just one small aspect can be
important.

As an aside, this test case also would have caught the bug fixed in
94c40909cc and by reverting that patch it
indeed fails.
2019-03-06 07:09:39 -05:00
James Shubin
de1691665f lang: funcs: Add live function stream test infrastructure
This adds the ability to test that functions return the expected
streams, and to model this behaviour over time. This is done via a
"timeline" which runs an ordered list of actions that can both push new
values into the function stream, and wait and expect particular outputs.
Hopefully this will make our function implementations more robust!
2019-03-06 07:09:39 -05:00
James Shubin
b1f93b40ae lang: funcs: Add runner pure func execution
This adds a function runner that runs pure functions. It will hopefully
be useful for speculative execution of functions for compile time
determination of types.
2019-03-05 11:42:33 -05:00
James Shubin
5e58251026 test: Improve govet log newline check
We don't match for log.Fatalf but we shouldn't really be using that.
2019-03-05 11:42:33 -05:00
James Shubin
4f4091a9bd engine: resources: Improve test case readability 2019-03-04 10:26:51 -05:00
James Shubin
e9fb41fdc8 test: shell: Fix rare breakage in load test
For some reason the load is occasionally zero. This broke the regexp.
Let's see if this ever happens with the other digits.
2019-03-04 10:16:21 -05:00
James Shubin
6b803656b2 engine: resources: Improve exec resource
The exec resource was an early addition to the project, and it was due
for some fixes and integration into our automated tests. This patch
fixes a number of issues, and makes it ready for more general use.
2019-03-04 10:16:21 -05:00
James Shubin
829741e2ac lang: Print a clear message on module import containing unused stmt
If you run an import, you only include everything that's part of a
scope. This includes, variables, classes, and functions. Anything else
should cause a compile error. This cleans up the error by adding a
String() method to each Stmt in our AST.
2019-02-28 09:35:13 -05:00
James Shubin
94c40909cc lang: funcs: Avoid erroneous empty message in readfile
Readfile had a bug where it sent an empty string on startup. This has
ben fixed, and it now waits until the file contents are ready before
sending a string.
2019-02-28 08:56:10 -05:00
James Shubin
95dab16e6e lang: funcs: Allow the len function to determine str length 2019-02-28 08:54:11 -05:00
James Shubin
c049413b47 examples: lang: Add is_debian and is_redhat family example
This is just the beginning.
2019-02-28 08:53:25 -05:00
Nicolas Charles
2d45f95501 engine: resources: print: Add RefreshOnly option
Add option RefreshOnly (default to false) on print ressource, to print only when
notified by other resource. When a print is RefreshOnly, it can't be grouped anymore.
2019-02-27 15:31:45 +01:00
Nicolas Charles
3cfc76b635 lang: funcs: Added a function to detect Debian and RedHat like systems 2019-02-26 18:13:34 +01:00
James Shubin
d88874845c test: shell: Improve load test
This might have failed once in travis because of a short timeout.
Hopefully if this happens again, we'll now know why.
2019-02-24 14:10:01 -05:00
James Shubin
5e38c1c8fe examples: Remove old hcl examples
The hcl frontend was removed a while back. Might as well remove these
examples too.
2019-02-24 14:10:01 -05:00
James Shubin
ae7ebeedd1 engine: resources: Add CheckApply event detection to resource tests
This adds the ability to wait with a timeout for CheckApply happenings
in a resource. This helps avoid unnecessary long sleeping and timing
guesses. This also adds a cleanup function to run at the end.
2019-02-24 14:10:01 -05:00
James Shubin
652b657809 resources: exec: Avoid possible deadlock race
Some of the early code I wrote probably wouldn't pass my own reviews
today. Here's one example of a rare deadlock that could sometimes occur
when a Process/CheckApply caused a shutdown, but the bufio tried to send
on a channel that nobody was going to read any more. Now we properly
unblock that send with a context.
2019-02-24 12:28:59 -05:00
James Shubin
62a6e0da1d misc: Add two test helpers
Hopefully these make testing and debugging easier!
2019-02-24 12:28:59 -05:00
James Shubin
0d0d48d9f6 test: Shell tests should use unified timeout command 2019-02-24 12:28:59 -05:00
James Shubin
ab5957f1e9 make: Clean up the Makefiles so the output is more elegant
This avoids printing erroneous messages when nothing is actually
happening.
2019-02-24 12:28:59 -05:00
James Shubin
463ba23003 util: Improve the sync primitives. 2019-02-24 12:28:59 -05:00
James Shubin
ccad6e7e1a test: Enable and fix up some more tests
An unstable engine probably masked some of these issues.
2019-02-24 12:28:59 -05:00
James Shubin
aa165b5e17 engine: Add the retry loop around Process
This adds back the retry loop around Process. This is done as a
separate commit so you can more easily see the logic of the retry magic
This commit is similar but different to the earlier commit adding retry
around Watch.
2019-02-24 12:28:59 -05:00
James Shubin
f06e87377c engine: Add limit delay before Process can run
This adds back the limit delay around Process.
2019-02-24 12:28:59 -05:00
James Shubin
4c3bf9fc7a engine: Add the retry loop around Watch
This adds back the retry loop around Watch. This is done as a separate
commit so you can more easily see the logic of the retry magic.
2019-02-24 12:28:59 -05:00
James Shubin
253ed78cc6 engine: Rewrite the core algorithm
The engine core had some unfortunate bugs that were the result of some
early design errors when I wasn't as familiar with channels. I've
finally rewritten most of the bad parts, and I think it's much more
logical and stable now.

This also simplifies the resource API, since more of the work is done
completely in the engine, and hidden from view.

Lastly, this adds a few new metaparameters and associated code.

There are still some open problems left to solve, but hopefully this
brings us one step closer.
2019-02-24 12:28:59 -05:00
James Shubin
4860d833c7 converger: Rewrite the converger module
I found a deadlock in the converger code, and I realized the code was
sufficiently bad that it needed a good clean up.
2019-02-24 12:28:59 -05:00
James Shubin
450d5c1a59 util: Add an easy ACK sync primitive 2019-02-24 12:28:59 -05:00
Toshaan Bharvani
88fcda2c99 lang: funcs: Added an uptime function
Signed-off-by: Toshaan Bharvani <toshaan@vantosh.com>
2019-02-24 12:20:58 -05:00
James Shubin
00db953c9f lang: funcs: funcgen: Clean up some small details
Some small changes were needed, here they are. Unfortunately this only
supports the `string` type at the moment.
2019-02-21 13:06:29 -05:00
Julien Pivotto
a0df4829a8 lang: Add more string functions, autogenerated
Signed-off-by: Julien Pivotto <roidelapluie@inuits.eu>
2019-02-21 17:50:06 +01:00
James Shubin
b0e1f12c22 test: Add expanders when running in travis
Hopefully this makes things more readable.
2019-02-20 09:35:31 -05:00
James Shubin
ee56155ec4 test: Split travis tests into three blocks
Our tests were taking near 50 minutes which kills them. This also makes
it easier to spot small issues faster.
2019-02-20 09:35:02 -05:00
Jeff Waugh
16d7c6a933 build: Fix macOS build
Add pkg-config to fix builds with augeas and libvirt on macOS.
2019-02-14 23:06:18 +11:00
Johan Bloemberg
f7a06c1da9 etcd: Connection options (socket file, ipv6)
- Allow unix domain socket to be used as client url
- Using ::1 as clienturl should not create default local ipv4 listener
- Add shell tests
2019-02-13 18:55:20 +01:00
James Shubin
4c8086977a engine: resources: file: Update the format string
The %s in the format string is not technically correct here.
2019-02-08 12:38:10 -05:00
James Shubin
b1f088e5fa engine: resources: Add a test running for testing individual resources
This adds a simulated engine that can run and test single resources. It
can't test all aspects and features that the engine supports, but is
probably pretty decent for testing the actual CheckApply and Watch
semantics. Be warned that it actually applies changes on your machine,
so please don't write tests that make undesirable changes.
2019-02-08 12:36:37 -05:00
James Shubin
1247c789aa lang: Remove unnecessary log package 2019-02-08 10:23:44 -05:00
Johan Bloemberg
749038c76d misc: Make build on macOS work 2019-02-08 00:14:17 +01:00
Johan Bloemberg
0a052494c4 misc: Add goimports dep 2019-02-08 00:14:17 +01:00
James Shubin
90fa83a5cf lang: funcs: core: Move world API functions
Some of the core functions interact with the remote "world" API. Move
them all into the same package.
2019-02-07 12:32:32 -05:00
James Shubin
4eaff892c1 lang: funcs: core: Rename core module files
More cleanup...
2019-02-07 12:19:59 -05:00
James Shubin
f368f75209 lang: funcs: core: Drop unnecessary core prefix from imports
This unbreaks the mcl bindata code. Of course we could change the parser
to allow this prefix, but this is cleaner. The packages still have a
core prefix, which it seems we could also remove, but this isn't
particularly important for anything.
2019-02-07 09:33:20 -05:00
Lander Van den Bulcke
04048b13ed lang: funcs: Add strings.split function
Signed-off-by: Lander Van den Bulcke <landervdb@inuits.eu>
2019-02-07 10:55:39 +01:00
Lander Van den Bulcke
5acc33c751 lang: funcs: Add tests for sqrt function
Signed-off-by: Lander Van den Bulcke <landervdb@inuits.eu>
2019-02-06 17:11:42 +01:00
James Shubin
b449be89a7 examples: Add uncommited nspawn example 2019-02-06 08:57:11 -05:00
Lander Van den Bulcke
dac019290d lang: funcs: Add sqrt function
Signed-off-by: Lander Van den Bulcke <landervdb@inuits.eu>
2019-02-06 14:32:13 +01:00
Julien Pivotto
bdc424e39d lang: Add to_lower and to_upper functions
Signed-off-by: Julien Pivotto <roidelapluie@inuits.eu>
2019-02-06 14:24:15 +01:00
Lander Van den Bulcke
10193a2796 make: Use gem --no-document instead of deprecated flags
Signed-off-by: Lander Van den Bulcke <landervdb@inuits.eu>
2019-02-06 12:02:10 +01:00
Julien Pivotto
2c9a12e941 docker: Update FROM to go:1.11
Signed-off-by: Julien Pivotto <roidelapluie@inuits.eu>
2019-02-06 10:48:24 +01:00
Felix Frank
8ba6c40f0c langpuppet: Fix Cli method invocations for wrapped GAPIs
Since the langpuppet GAPI creates fresh new CliContext objects,
it has to make sure to provide the original parent context, because
the child GAPIs expect to be able to access its data.
2019-02-05 16:34:55 +01:00
James Shubin
bbfeb49cdf examples: Add more examples and clean up some 2019-02-04 05:03:37 -05:00
James Shubin
f61e1cb36d examples: Add missing mcl files
I forgot to add these, sorry.
2019-02-03 09:58:04 -05:00
James Shubin
4a3e2c3611 engine: nspawn: Add an nspawn example with an improved exec
This adds the cwd fields to exec, better error messages to svc (which is
nested in nspawn) and a fancier nspawn example!
2019-02-01 09:44:55 -05:00
James Shubin
81faec508c integration: Avoid duplicate events from recwatch 2019-02-01 07:58:38 -05:00
James Shubin
9966ca2e85 examples: Improve dynamic cpus virt example 2019-02-01 07:58:38 -05:00
James Shubin
35c26f9ee5 engine: resources: virt: Clean up virt resource for lang 2019-02-01 07:58:38 -05:00
James Shubin
b5e29771ab lang: funcs: Add a trim space function to the new strings module 2019-02-01 07:00:05 -05:00
James Shubin
f5f09d3640 lang: funcs: Add str2int example function
We might want to move this into a real module eventually.
2019-02-01 06:59:07 -05:00
James Shubin
5a531b7948 lang: funcs: Add a new readfile function
This adds a new function that reads files from the local host.
2019-02-01 05:20:22 -05:00
James Shubin
f716a3a73b lang: funcs: Rename template functions to remove periods
Due to a limitation in the template library, we need to rename some
functions. It's probably worth looking into modifying this library or
finding an alternate version.
2019-02-01 03:58:02 -05:00
James Shubin
ce8c8c8eea engine: resources: Fix a small typo in error message 2019-02-01 03:49:08 -05:00
James Shubin
fc48fda7e5 engine: resources: Fix a possible panic on closed channel
I don't know how often it happens, but we should catch it.
2019-02-01 03:48:24 -05:00
James Shubin
78936c5ce8 examples: lang: Update examples to fix imports and port from yaml
Some small fixes that are useful for demos!
2019-02-01 03:47:18 -05:00
Kevin Kuehler
5d0efce278 engine: lang: util: Kill race in socketset
After some investigation, it appears that SocketSet.Shutdown() and
SocketSet.Close() are not synchronous operations. The sendto system call
called in SocketSet.Shutdown() is not a blocking send. That means there
is a race in which SocketSet.Shutdown() sends a message to a file
descriptor to unblock select, while SocketSet.Close() will close the
file descriptor that the message is being sent to. If SocketSet.Close()
wins the race, select is listening on a dead file descriptor and will
hang indefinitely.

This is fixed in the current master by putting SocketSet.Close() inside
of the goroutine in which data from the socket is being received. It
relies on SocketSet.Shutdown() being called to terminate the goroutine.
While this works most of the time, there is a race here. All the
goroutines can also be terminated by a closeChan. If the goroutine
receives an event (thus unblocking select) and then closeChan is
triggered, both SocketSet.Shutdown() and SocketSet.Close() race, leading
to undefined behavior.

This patch ensures the ordering of the two function calls by pulling
them both out of the goroutine and separating them with a WaitGroup.

Co-authored-by: James Shubin <james@shubin.ca>
2019-01-22 20:59:17 -08:00
Kevin Kuehler
0c17a0b4f2 util: Add TestShutdown to socketset
Test to ensure that SocketSet is nonblocking and will close when
SocketSet.Shutdown() is called. Create a SocketSet that will never
receive any data and leave it running in a goroutine with a WaitGroup
for a second. If Shutdown is working correctly, the goroutine will be
terminated after the timer expires.
2019-01-22 20:59:17 -08:00
Kevin Kuehler
3f396a7c52 lang: funcs: Add cpucount fact
Adds a CPU count fact, that can be used to determine how many CPUs are
presently on the machine and ready for use (online). We get this by
reading from a netlink socket to the kernel, and the kernel sends us
uevents when CPUs are added, removed, and brought online or offline.
Whenever one of these events are received, we look in sysfs to update
the fact's Stream with the number of online CPUs.
2019-01-22 20:59:16 -08:00
Kevin Kuehler
8697f8f91f util: Libify socketset
Add the ReceiveBytes, ReceiveNetlinkMessage, and ReceiveUEvent methods.
This is because not everything passed through a netlink socket cannot
reliably be parsed using the ParseNetLinkMessage function.

With the ReceiveUEvent method, we add support for "uevent" kernel
events, which updates us about the state of devices currently on the
system. To make using this method easier, we add a UEvent struct, that
has the action (what event), Devpath (where the device lives in /proc or
/sysfs), and Subsystem (what subsystem this event belows to).
2019-01-22 20:59:16 -08:00
Kevin Kuehler
06c67685f1 util: Move socketset from net resource to util
Prepare the socketset api to be used outside of the scope the net
resource.
2019-01-22 20:59:11 -08:00
James Shubin
dc2e7de9e5 engine: resources: pkg: Clarify that correct state is newest
I accidentally typed "latest" which got me confused why everything was
broken. Surprised it didn't error earlier anyways.
2019-01-21 04:28:34 -05:00
James Shubin
db1dbe7a27 lang: Edges should allow lists of strings
This continues the earlier patch that allowed resource names to be lists
of strings so that edges can now allow the same. This also includes a
new fancy test!
2019-01-20 17:27:40 -05:00
James Shubin
d6bbb94be5 lang: test: Add a new giant test infra for matching static output
This greatly expands our test infra to allow us to drop in mcl tests and
look at their resource graph output. The only downside is that this only
runs the function engine once, so if the function graph would be
constantly changing over time, then this is not a good fit here.
2019-01-20 17:27:40 -05:00
James Shubin
e3b4c0aee3 test: Fix a small copy pasta typo 2019-01-20 17:27:40 -05:00
James Shubin
a1fbe152bb lang: unification: Fix up small typos in example code 2019-01-20 04:22:05 -05:00
James Shubin
9d28ff9b23 lang: unification: Catch unification error on typed var expr
This was similar to the typed if expr error.
2019-01-20 04:19:39 -05:00
James Shubin
43f0ddd25d lang: unification: Catch unification error on typed if expr
I found a case where we had two missing unification rules. Now fixed in
the previous commits, and including this test to show I'm responsible.
I've added the same test in two locations for redundancy and as an
example.
2019-01-20 04:19:39 -05:00
James Shubin
7a28b00d75 lang: If expression was missing two invariants
I forgot to ensure that the type of the final expression matched the
type of each of the branches. It's rare, but possible for this to occur.
Luckily, this never would have caused a panic, because the func engine
would have caught the issue anyways, but it's still better we catch it
here first!
2019-01-20 04:02:54 -05:00
James Shubin
32e29862f2 lang: Check that set type matches actual expression
I forgot to include these two invariants which are occasionally
necessary, although in most cases they're necessary to prevent incorrect
code from getting past unification. In any case, they would have been
caught by the engine.
2019-01-20 04:02:54 -05:00
James Shubin
6c5c38f5a7 lang: unification: Allow err string comparisons in tests
Let's improve our test infra to make it more capable. It's important to
catch we failed for the _right_ reason so as to not mask the wrong
errors.
2019-01-20 03:39:02 -05:00
James Shubin
2da7854b24 lang: unification: Add logging to make capturing errors easier
This makes building new tests easier.
2019-01-20 03:39:02 -05:00
James Shubin
6d0c5ab2d5 lang: unification: Add missing return to exit early
This exits the test early, since we don't need to continue.
2019-01-20 03:39:02 -05:00
James Shubin
9398deeabc etcd: Workaround a nil ptr bug
A clean re-write of this etcd code is needed, but until then, this
should hopefully workaround the occasional test failures. In practice I
don't think anyone has every hit this bug.
2019-01-17 20:07:24 -05:00
James Shubin
bf63d2e844 engine: graph: Avoid a possible panic sending on a closed channel
It's plausible that we send on a closed channel if we're running a back
poke and it tries to send a poke on something that has already closed.
If it detects this condition, it will exit.

Unfortunately, it's not clear if the wait group will protect this case,
but hopefully this will hold us until we can re-write the needed parts
of the engine.
2019-01-17 20:05:49 -05:00
James Shubin
b808592fb3 engine: Work around bad timestamp panic
Occasionally when a back poke happens downstream of an upstream vertex
which has already exited, it could get back poked, which would cause a
panic. This moves the deletion of the state struct until the entire
graph has completed so that it won't panic. It doesn't matter if a back
poke is lost, we're shutting down or pausing, and in this scenario it
can be lost.
2019-01-17 20:05:49 -05:00
James Shubin
e2296a631b engine: event: Switch events system to use simpler structs
Pass around pointers of things now. Also, naming is vastly improved and
clearer.
2019-01-17 20:04:17 -05:00
James Shubin
e20555d4bc test: Don't be unnecessarily noisy in this test
This is confusing if you're looking for an error in the test.
2019-01-17 19:33:35 -05:00
James Shubin
b89e2dcd3c test: Add a three host variant of the empty etcd test 2019-01-17 19:21:56 -05:00
James Shubin
165d11b2ca test: Rename t8 to be more descriptive 2019-01-17 19:21:56 -05:00
James Shubin
d4046c0acf test: Enable t8 to test for two host etcd clusters
I can't remember why we disabled this, so let's put it back. There's
still one rare etcd race, but hopefully it doesn't fail too much until
we fix it.
2019-01-17 19:21:56 -05:00
James Shubin
88498695ac test: Add a semaphore shell test
This test tests new language features and as a fan in-out graph.
2019-01-17 19:21:56 -05:00
James Shubin
354a1c23b0 engine: graph: Prevent converged timeout of dirty res
Somewhere after the engine re-write we seem to have regressed and
converge early even if some resource is dirty. This adds an additional
timer so that we don't start the individual resource converged countdown
until our state is okay.
2019-01-17 18:46:00 -05:00
Kevin Kuehler
34550246f4 lang: Add debug flag and Logf to fact init struct 2019-01-17 18:12:45 -05:00
Jonathan Gold
db1cc846dc test: Ensure gometalinter is available 2019-01-15 20:24:37 -05:00
Jonathan Gold
74484bcbdf make: deps: Only install gometalinter on CI/CD servers 2019-01-15 20:23:24 -05:00
James Shubin
d5ecf8ce16 engine: Fix typos 2019-01-12 15:03:03 -05:00
James Shubin
b1ffb1d4a4 lang: Add autoedge and autogroup meta params to mcl
These weren't yet exposed in mcl. They're now available under the same
Meta namespace as the normal meta param structs. Even though they live
as a separate trait, they should be exposed together for a consistent
interface in mcl. If autoedge or autogroup ever grow additional params,
we can always add: `Meta:autoedge:something` to break it down further.
2019-01-12 13:16:39 -05:00
James Shubin
451e1122a7 lang: Refactor the res metaparams helper
We can do all the actions without returning anything but an error.
2019-01-12 12:34:07 -05:00
James Shubin
10dcf32f3c lang: Allow a list of strings in the resource name
This adds a core looping construct by allowing a list of names to build
a resource. They'll all have the same parameters, but they'll
intelligently add the correct list of edges that they'd individually
create.

Constructs like these are one reason we do NOT have actual looping
functionality in the language, and it should stay that way.
2019-01-12 11:54:02 -05:00
James Shubin
7f1477b26d lang: Add a placeholder "ExprAny" expression for unification hacks
Instead of adding complexity to the unification engine, we can add a
fake placeholder expression that is unreachable by the AST, but used for
unification so that we can ensure a "wrap" invariant has some contents.

Ideally we'd improve the unification engine, but we'll leave that for
the future, and it's easy to revert this one commit in the future.
2019-01-12 11:45:53 -05:00
James Shubin
33b68c09d3 lang: Refactor edges helper method 2019-01-12 11:45:53 -05:00
James Shubin
7ec48ca845 lang: Refactor resource creation into a helper method 2019-01-12 11:45:53 -05:00
James Shubin
5c92cef983 docs: Add sub categories to the language guide
Hopefully this makes the longer sections easier to read.
2019-01-12 11:45:53 -05:00
James Shubin
75eba466c6 travis: Clean up my grammar
What was I thinking?
2019-01-11 04:38:12 -05:00
James Shubin
ad30737119 lang: Add meta parameter parsing to resources
Now we can actually specify metaparameters in the resources!
2019-01-11 04:13:13 -05:00
James Shubin
8e0bde3071 lang: Move capitalized res identifier into parser
This gives us more specificity when trying to match exactly.
2019-01-11 02:57:39 -05:00
James Shubin
7d641427d2 test: Fix golang cache regression
Golang decided to change the GOCACHE behaviour in newer versions of `go
test`. This changes our tests to use the new approach.

For users using a local `.envrc`, you might want to add:

GOFLAGS="-count=1"

Which is supposed to fix this problem for local tests.

More information is available in: https://github.com/golang/go/issues/29378
2019-01-10 20:41:10 -05:00
James Shubin
3b62beed26 travis: Print debug info to catch travis regressions 2019-01-10 18:23:11 -05:00
James Shubin
2d3cf68261 travis: Workaround another broken apt repo
This works around another travis NO_PUBKEY regression.
2019-01-10 18:22:45 -05:00
Vincent Membré
7d6080d13f engine: resources: exec: Use WatchShell in Exec resource when needed instead of Shell 2019-01-03 10:28:22 +01:00
James Shubin
e3eefeb3fe engine: resources: pkg: Implement the CompatibleRes interface
This signals to an interested consumer that two or more compatible
resources can be merged safely. This is so that we can avoid the
"duplicate resource" design problem that puppet had.

To test this, you can run:

./mgmt run --tmp-prefix lang --lang 'pkg "cowsay" { state => "installed", } pkg "cowsay" { state => "newest", }'

which should work.
2018-12-29 02:54:55 -05:00
James Shubin
f10dddadd6 lang: Handle merging of compatible resources properly
The duplicate resource problem that puppet had should now be correctly
solved in mgmt.
2018-12-29 02:51:09 -05:00
James Shubin
d166112917 engine: Add an interface for compatible resources
This also adds utility functions for merging and improved comparing.
2018-12-29 02:46:43 -05:00
James Shubin
8ed5c1bedf engine: Add a resource copy interface and implementation
If we want to copy an entire resource, we should use this helper method.
2018-12-29 02:42:02 -05:00
James Shubin
4489076fac engine: Add setters for the trait interfaces
Turns out it's useful to wholesale set the entire struct.
2018-12-29 01:16:38 -05:00
James Shubin
bdc33cd421 lang: Validate the edge field names in our resources
Validate these early instead of waiting for this to be caught during
output generation.
2018-12-29 00:18:10 -05:00
James Shubin
889dae2955 lang: Improve sub testing
This makes individual sub tests from the table easier to run.
2018-12-29 00:16:35 -05:00
James Shubin
9ff21b68e4 engine: resources: pkg: Simplify state check
Refactor this code.
2018-12-28 20:33:51 -05:00
James Shubin
a69a7009f8 engine: resources: pkg: Replace state strings with constants
This helps avoid typos, and gives us something we can export in the
future.
2018-12-28 20:32:23 -05:00
James Shubin
d413fac4cb engine: resources: pkg: Remove old Compare method
This was legacy code. Get rid of it.
2018-12-28 20:06:00 -05:00
James Shubin
246ecd8607 engine: resources: cron: Fix typo in error message 2018-12-28 20:00:14 -05:00
James Shubin
22105af720 lang: test: Add a test of duplicate resource generation
These two cases should be allowed in our language. This is something
that puppet got wrong, and hopefully this makes writing modules more
sane in mcl, since two modules both depending on a "cowsay" package
won't cause compile errors.

This only checks the language. The de-duplication is done there. We
don't currently have a check for this in the engine. (We should!)
2018-12-28 18:44:07 -05:00
James Shubin
880c4d2f48 lang, util: Tests that depend on the fs should be sorted
This ensures they're deterministic on any file system.
2018-12-28 18:00:08 -05:00
Jonathan Gold
443f489152 etcd: Add more test cases to TestEtcdCopyFs0 2018-12-22 04:47:49 -05:00
Jonathan Gold
39fdfdfd8c etcd: Add TestEtcdCopyFs0
This commit adds a new test to etcd/fs/fs_test.go that performs the same
actions (with some new cases) as TestFs2 and TestFs3, but allows us to
add more test cases as needed.
2018-12-22 04:46:58 -05:00
James Shubin
96dccca475 lang: Add module imports and more
This enables imports in mcl code, and is one of last remaining blockers
to using mgmt. Now we can start writing standalone modules, and adding
standard library functions as needed. There's still lots to do, but this
was a big missing piece. It was much harder to get right than I had
expected, but I think it's solid!

This unfortunately large commit is the result of some wild hacking I've
been doing for the past little while. It's the result of a rebase that
broke many "wip" commits that tracked my private progress, into
something that's not gratuitously messy for our git logs. Since this was
a learning and discovery process for me, I've "erased" the confusing git
history that wouldn't have helped. I'm happy to discuss the dead-ends,
and a small portion of that code was even left in for possible future
use.

This patch includes:

* A change to the cli interface:
You now specify the front-end explicitly, instead of leaving it up to
the front-end to decide when to "activate". For example, instead of:

mgmt run --lang code.mcl

we now do:

mgmt run lang --lang code.mcl

We might rename the --lang flag in the future to avoid the awkward word
repetition. Suggestions welcome, but I'm considering "input". One
side-effect of this change, is that flags which are "engine" specific
now must be specified with "run" before the front-end name. Eg:

mgmt run --tmp-prefix lang --lang code.mcl

instead of putting --tmp-prefix at the end. We also changed the GAPI
slightly, but I've patched all code that used it. This also makes things
consistent with the "deploy" command.

* The deploys are more robust and let you deploy after a run
This has been vastly improved and let's mgmt really run as a smart
engine that can handle different workloads. If you don't want to deploy
when you've started with `run` or if one comes in, you can use the
--no-watch-deploy option to block new deploys.

* The import statement exists and works!
We now have a working `import` statement. Read the docs, and try it out.
I think it's quite elegant how it fits in with `SetScope`. Have a look.
As a result, we now have some built-in functions available in modules.
This also adds the metadata.yaml entry-point for all modules. Have a
look at the examples or the tests. The bulk of the patch is to support
this.

* Improved lang input parsing code:
I re-wrote the parsing that determined what ran when we passed different
things to --lang. Deciding between running an mcl file or raw code is
now handled in a more intelligent, and re-usable way. See the inputs.go
file if you want to have a look. One casualty is that you can't stream
code from stdin *directly* to the front-end, it's encapsulated into a
deploy first. You can still use stdin though! I doubt anyone will notice
this change.

* The scope was extended to include functions and classes:
Go forth and import lovely code. All these exist in scopes now, and can
be re-used!

* Function calls actually use the scope now. Glad I got this sorted out.

* There is import cycle detection for modules!
Yes, this is another dag. I think that's #4. I guess they're useful.

* A ton of tests and new test infra was added!
This should make it much easier to add new tests that run mcl code. Have
a look at TestAstFunc1 to see how to add more of these.

As usual, I'll try to keep these commits smaller in the future!
2018-12-21 06:22:12 -05:00
James Shubin
948a3c6d08 gapi: Add a bytes helper
Use bytes directly if we've got them.
2018-12-20 21:21:30 -05:00
James Shubin
dc13d5d26b util: Add some useful path parsing functions
These two are useful for looking at path prefixes and rebasing the paths
onto other paths.
2018-12-20 21:21:30 -05:00
James Shubin
aae714db6b lang: Add a top-level stmt safety method
This adds a new method to the *StmtProg that lets us determine if the
prog contains only what is necessary for a scope and nothing more. This
is useful because that is exactly what is produced when doing an import.
With this detection method, we can know if a module contains dead code
that might mislead the user into thinking it will get run when it won't.
2018-12-20 21:21:30 -05:00
James Shubin
a7c9673bcf lang: Improve empty scope and output
For some reason these were unnecessary methods on the structs, even when
those structs contained nothing useful to offer.
2018-12-20 21:21:30 -05:00
James Shubin
3d06775ddc lang: Add some lambda function parsing and tests
Part of this isn't fully implemented, but might as well get the tests
running.
2018-12-20 21:21:30 -05:00
James Shubin
48beea3884 test: Clean up and improve golang tests
This adds some consistency to the tests and properly catches difficult
scenarios in some of the lexparse tests.
2018-12-20 21:21:30 -05:00
James Shubin
958d3f6094 lang: Add beginning of user defined functions
This adds the lexer, parser and struct basics for user defined
functions. It's far from finished, but it's good to get the foundation
started.
2018-12-20 21:21:30 -05:00
James Shubin
08f24fb272 lang: Add a URL result to the import name parser
This is meant to be useful for the downloader. This will probably get
more complicated over time, but for now the goal is to have it simple
enough to work for 80% of use cases.
2018-12-20 21:21:30 -05:00
James Shubin
07d57e1a64 git: Ignore some WIP files that won't get tracked in git 2018-12-20 21:21:30 -05:00
James Shubin
cd7711bdfe gapi: Add a prefix variable in case we want to namespace on disk
This could get passed through to use as a module download path.
2018-12-20 21:21:30 -05:00
James Shubin
433ffa05a5 bindata: Add infrastructure for building core mcl files
This should prepare us so that we can build native mcl code alongside
the core *.go files which we already have. This includes a single mcl
file that is used as a placeholder so that the build doesn't fail if we
don't have any mcl files in the core/ directory. It will get ignored
automatically.
2018-12-20 21:21:30 -05:00
James Shubin
046b21b907 lang: Refactor most functions to support modules
This is a giant refactor to move functions into a hierarchial module
layout. While this isn't entirely implemented yet, it should work
correctly once all the import bits have landed. What's broken at the
moment is the template function, which currently doesn't understand the
period separator.
2018-12-20 21:21:30 -05:00
James Shubin
c32183eb70 lang: Tidy up grouping of lexer tokens in the parser
Just some small cleaning.
2018-12-20 21:21:30 -05:00
James Shubin
73b11045f2 lang: Add lexing/parsing of import statements
This adds the basic import statement, and its associated variants. It
also adds the import structure which is the result of parsing.
2018-12-20 21:21:30 -05:00
James Shubin
57ce3fa587 lang: Allow matching underscores in some of the identifier's
This allows matching underscores in some of the identifier's, but not
when they're the last character.

This caused me to suffer a bit of pain tracking down a bug which turned
out to be in the lexer. It started with a failing test that I wrote in:

974c2498c4

and which followed with a fix in:

52682f463a

Glad that's fixed!
2018-12-20 21:21:30 -05:00
James Shubin
a26620da38 lang: Add resource specific tokens in lexer and parser
This adds some custom tokens for the lexer and parser so that resources
can have colons in their names.
2018-12-20 21:21:30 -05:00
James Shubin
86b8099eb9 lang: Add import spec parsing and tests
This adds parsing of the upcoming "import" statement contents. It is the
logic which determines how an import statement is read in the language.
Hopefully it won't need any changes or additional magic additions.
2018-12-20 21:21:30 -05:00
James Shubin
c8e9a100a6 lang: Support lexing and parsing a list of files with offsets
This adds a LexParseWithOffsets method that also takes a list of offsets
to be used if our input stream is composed of multiple io.Readers
combined together.

At the moment the offsets are based on line count instead of file size.
I think the latter would be preferable, but it seems it's much more
difficult to implement as it probably requires support in the lexer and
parser. That improved solution would probably be faster, and more
correct in case someone passed in a file without a trailing newline.
2018-12-20 21:21:30 -05:00
James Shubin
a287f028d1 lang: Detect sub tests with the same name
This detects identically named tests and fails the test in such a
scenario to prevent confusion.
2018-12-20 21:21:30 -05:00
James Shubin
cf50fb3568 lang: Allow dotted identifiers
This adds support for dotted identifiers in include statements, var
expressions and function call expressions. The dotted identifiers are
used to refer to classes, bind statements, and function definitions
(respectively) that are included in the scope by import statements.
2018-12-20 21:21:30 -05:00
James Shubin
4c8193876f util: Add a UInt64Slice and associated sorting functionality.
This adds an easy to sort slice of uint64's and associated functionality
to sort a list of strings by their associated order in a map indexed by
uint64's.
2018-12-20 21:21:30 -05:00
James Shubin
158bc1eb2a lang: Add an Apply iterator to the Stmt and Expr API
This adds a new interface Node which must implement the Apply method.
This method traverse the entire AST and applies a function to each node.
Both Stmt and Expr must implement this.
2018-12-20 21:21:30 -05:00
James Shubin
3f42e5f702 lang: Add logging and debug info via a new Init method
This expands the Stmt and Expr interfaces to add an Init method. This
is used to pass in Debug and Logf values, but is also used to validate
the AST. This gets rid of standalone use of the "log" package.
2018-12-20 21:21:30 -05:00
Tom Payne
75633817a7 etcd: Ensure that fs.Fs implements afero.Fs 2018-12-20 21:19:55 -05:00
Tom Payne
83b00fce3e etcd: Add Lchown (returns ErrNotImplemented) 2018-12-20 21:19:55 -05:00
Tom Payne
38befb53ad etcd: Add Chown (returns ErrNotImplemented) 2018-12-20 21:19:55 -05:00
Kevin Kuehler
d0b5c4de68 util: Patch CopyFs and add tests
Fix CopyFs bug that resulted in a flattened destination directory.
Added tests catch this bug, and ensure the data is in fact copied
to the destination directory.
2018-12-20 12:15:06 -08:00
James Shubin
1b68845b00 test: Fix up token vet test
I forgot some of the cases to catch earlier.
2018-12-19 22:24:20 -05:00
James Shubin
a7bc72540d util: Fix small linting error
Woops!
2018-12-19 12:29:44 -05:00
James Shubin
27ac7481f9 test: Increase the vet testing for irregular strings
Catch some inconsistent comments to keep things neat. Hey, anything we
can automate, we do :)
2018-12-19 06:52:23 -05:00
James Shubin
9bc36be513 util: Add a test for CopyFs
This adds a test case for the standalone CopyFs function, and an easy to
use test case infra.
2018-12-19 06:51:05 -05:00
James Shubin
e62e35bc88 util: Improve the test helper function and add a better one
This should help us write tests that use unique physical directories
inside the directory tree.
2018-12-19 06:10:48 -05:00
James Shubin
bd80ced9b2 util: Add an fs helper and a test helper 2018-12-17 12:10:09 -05:00
Jonathan Gold
bb2f2e5e54 util: Add PathSlice type that satisfies sort.Interface
This commit adds a []string{} type alias named PathSlice, and the
Len(), Swap(), and Less() methods required to satisfy sort.Interface.
Now you can do `sort.Sort(util.PathSlice(foo))` where foo is a slice
of paths. It will be sorted by depth in alphabetical order.
2018-12-17 01:14:54 -05:00
James Shubin
b1eb6711b7 engine: resources: Work around a subtle embedded res bug
This is a subtle issue that was found that caused a panic. This should
solve things for now, but it would be wise to build embedded or
composite resources sparingly until we we're certain this would work the
way we wanted for all scenarios.
2018-12-16 16:07:42 -05:00
Jonathan Gold
da0ffa5e56 engine: resources: cron: Add auto edges from SvcRes 2018-12-16 15:12:58 -05:00
Felix Frank
68ef312233 gitignore: Ignore vim swap files 2018-12-16 13:41:21 -05:00
Felix Frank
9fefadca24 docs: Explain the langpuppet interface and function 2018-12-16 13:35:47 -05:00
James Shubin
e14b14b88c engine: resources: svc: Add symmetric closing
This improves some of the closing in the svc resource. This still needs
lots of improvements, and it's sort of terrible because it was some very
early code written.
2018-12-16 08:27:26 -05:00
James Shubin
d5bfb7257e engine: resources: file: Require paths to be absolute
This is a requirement of our file resource, so we should validate this
and clearly express it in the documentation.
2018-12-16 07:24:07 -05:00
Jonathan Gold
8282f3b59c engine: resources: cron: Add lang examples 2018-12-15 11:01:05 -05:00
Jonathan Gold
dbf0c84f0b engine: resources: cron: Add support for user session timers 2018-12-15 10:47:35 -05:00
Jonathan Gold
a5977b993a engine: util: Add EdgeCombiner() for combining auto edges 2018-12-15 10:47:35 -05:00
Jonathan Gold
27df3ae876 engine: resources: cron: Add a systemd-timer resource 2018-12-15 10:47:35 -05:00
Felix Frank
a49d07cf01 gapi: langpuppet: Add initial implementation
This new entrypoint allows graph generation from both a Puppet manifest
and a piece of mcl code. The GAPI implementation wraps the two existing
GAPIs.
2018-12-15 03:43:15 +01:00
Jonathan Gold
28f343ac50 engine: resources: svc: Use dbus session bus for user session svc
This patch adds a util function, SessionBusUsable, that makes and returns
a new usable dbus session bus. If the svc bool session is true, the resource
will use a bus created with that function.
2018-12-14 00:16:21 -05:00
Jonathan Gold
4297a39d03 engine: resources: group: Make group edgeable
This adds the edgeable trait to the group resource and adds an
AutoEdges method which returns nil, nil. These changes are necessary
to allow UserRes to make autoedges to GroupRes.
2018-12-13 23:01:42 -05:00
Jonathan Gold
bd996e441c etcd: Use mgmt backend for fs tests 2018-12-11 18:11:45 -05:00
Jonathan Gold
086a89fad6 etcd: Use source filepath base in CopyFs destination path
This patch corrects the destination path in CopyFs to use the source's
base filepath, instead of the entire source path. Now copying /foo/bar
to /baz results in /baz/bar instead of /baz/foo/bar. This commit also adds
a test to verify this behaviour.
2018-12-11 02:20:11 -05:00
Michael Lesko-Krleza
70ac38e66c test: Increase test coverage for graphsync
This patch is an addition to graphsync_test.go, which increases the test
coverage from 72.4% to 72.9%.
2018-12-11 02:02:33 -05:00
James Shubin
d990d2ad86 travis: Bump to golang 1.10
This requires breaking changes in gofmt. It is hilarious that this was
changed. Oh well. This also moves to the latest stable etcd. Lastly,
this changes the `go vet` testing to test by package, since the new go
vet changed how it works and now fails without this change.
2018-12-11 01:46:17 -05:00
Jonathan Gold
56db31ca43 engine: resources: file: Add shell test for source field 2018-12-10 22:08:59 -05:00
Jonathan Gold
b902e2d30b engine: resources: file: Fix bug preventing use of source field
This patch fixes a previously undiscovered bug which prevented
the use of the source field in the file resource. CheckApply was
returning early if obj.Content was nil. It is also necessary to
check that obj.Source is empty before returning, otherwise
syncCheckApply never runs.
2018-12-10 22:08:59 -05:00
Jonathan Gold
d2bab32b0e engine: resources: packagekit: Fix dbus addmatch rule
I broke packagekit with commit 299080f5 due to a missing equals sign
in the DBus AddMatch rule. This commit adds the necessary equals sign.
2018-12-09 11:12:48 -05:00
Jonathan Gold
b2d726051b travis: Build on Xenial
Builds were failing on Trusty due to broken GPG keys, and upgrading
the build environment to Xenial Xerus solves the problem.
2018-12-04 20:27:47 -05:00
Jonathan Gold
8e25667f87 engine: resources: net: test: Add shell test for net resource
This patch adds a shell test for net, which creates a dummy interface
and runs mgmt to bring it up and assign it with an address. It then
checks if the state was applied correctly. Finally, it runs mgmt again
to bring the interface down, and tests that it comes down and stays
down.
2018-12-04 17:12:57 -05:00
Jonathan Gold
9b5c4c50e7 engine: resources: net: Allow addr without gateway
In some scenarios it is desirable to set the addrs and gateway
independently, i.e. if a default gateway is already set on
the machine. This patch removes the requirement to set them
together.
2018-12-04 17:12:57 -05:00
Jonathan Gold
d2ce70a673 puppet: Fix error message when puppet conf copy fails
This commit adds the missing config file location to the error
message.
2018-12-04 16:58:40 -05:00
Felix Frank
9db0fc4ee4 make: Speed up the build by skipping gem docs
Per default, the Ruby gems renerate documentation in two distinct formats
during installation. By passing --no-ri and --no-rdoc, gem is instructed
to skip this step for both formats.

If the user needs documentation for any of the gems after all, they can
manually generate the docs themselves.
2018-12-04 16:56:23 -05:00
Felix Frank
9ed830bb81 make: Remove spurious dependency package 'rubygems' for Debian-like systems
On Ubuntu, the apt-get install call to ruby, ruby-devel, and rubygems will
fail because there is no "rubygems" package in Ubuntu.

In Debian, this package is virtual only. In both cases, the ruby package
is sufficient. (See also https://packages.debian.org/jessie/rubygems)
2018-12-04 16:55:46 -05:00
James Shubin
4e42d9ed03 travis: Work around broken travis NO_PUBKEY error
W: GPG error: https://packagecloud.io/rabbitmq/rabbitmq-server/ubuntu trusty InRelease: The following signatures couldn't be verified because the public key is not available: NO_PUBKEY F6609E60DC62814E
E: The repository 'https://packagecloud.io/rabbitmq/rabbitmq-server/ubuntu trusty InRelease' is not signed.
N: Updating from such a repository can't be done securely, and is therefore disabled by default.
N: See apt-secure(8) manpage for repository creation and user configuration details.
2018-12-04 16:52:02 -05:00
James Shubin
4c93bc3599 test: Add doc note about skipping docker tests
This is useful if you don't have docker running, since otherwise it
causes all the tests to fail.
2018-12-03 23:55:20 -05:00
Jonathan Gold
7c817802a8 engine: resources: net: test: Add some go tests
This patch adds go tests for NetRes.unitFileContents(), socketSet.fdSet(),
and socketSet.nfd(), in the net resource.
2018-12-03 23:42:42 -05:00
Jonathan Gold
de90b592fb lang: Fix error message format strings
This commit replaces %s with %d in two error messages, where the
argument is an integer, not a string.
2018-12-03 19:27:35 -05:00
Jonathan Gold
b9d0cc2e28 etcd: Fix deploy transaction error message
This commit removes an unused argument from the error format string.
2018-12-03 19:26:18 -05:00
James Shubin
0ec00fe57f make: Improve release pipeline
Hopefully this makes releases a little better for users.
In particular, this avoids listing old build artifacts in the SHA256SUMS
files when we make new releases, and users can now download them
directly.

Now to make a release you run: `make tag && make release`.
After the first make session ends, you'll have a new tag released
publicly, and then during the second make session, the release target
will notice this new tag, build some assets, and upload them!
2018-11-30 19:08:53 -05:00
Jonathan Gold
80931e1cb4 make: Release pipeline
This commit adds new make targets for rpm, deb, and pacman packages.
It also adds a phony target that uploads tarballs of the packages,
along with their signed (and unsigned) checksums to the github release
page. Once the current commit is tagged as a release, run `make release`
to build the packages and upload them to github.
2018-11-30 04:53:51 -05:00
James Shubin
cc02e96a13 engine: resources: Add nodocker build tag
Make it easy to disable building docker which is enormous.
2018-11-29 08:22:05 -05:00
Jonathan Gold
51ec91dd16 engine: resources: docker: Add a docker container resource 2018-11-29 08:14:07 -05:00
James Shubin
916a92c3d8 vendor: Add vendored docker modules with out of tree fix
The docker project absurdly *copies* all of the dependencies into the
vendor/ directory instead of using git submodules or avoiding
unnecessary vendoring entirely. We manually remove these changes until
they learn to use tools how they're intended.

As an aside, we recommend using a more intelligent, modern tool like
systemd-nspawn instead.
2018-11-29 08:14:07 -05:00
James Shubin
5431bfdc29 test: Improve commit message tests 2018-11-24 04:42:50 -05:00
Jonathan Gold
43b5b4f5a4 build: Add rubygems to make deps target
cffdb06 adds a linter for markdown which requires rubygems.
This commit adds the dependency to the make target.
2018-10-30 17:16:50 -04:00
James Shubin
f342e06ef0 readme: Add Liberapay link to README 2018-06-21 19:21:41 -04:00
James Shubin
81bb87f4cd test: Add a test to ensure the parser doesn't have any conflicts
Our grammar shouldn't be ambiguous, and it makes sense to test this.
2018-06-18 16:06:23 -04:00
James Shubin
c4b97fadcc lang: Update map type definition to include a prefix
It turns out that some planned additions to the parser make it so that
the map type definition can be ambiguous. As a result, this patch
updates the definition so that the map definition is not confused with
an open curly bracket anywhere.

Thanks to pestle and stbenjamin for their help understanding yacc!
2018-06-18 16:06:23 -04:00
James Shubin
05f6ba7297 lang: Add partial recursive support/detection to class
This adds the additional bits onto the class/include statements to
support or detect class recursion. It's not currently supported, but
I figured I'd commit the detection code as a variant of the recursion
implementation, since I think this is correct, and it was a bit tricky
for me to get it right.
2018-06-17 17:35:34 -04:00
James Shubin
c62b8a5d4f lang: Add class and include statements
This adds support for the class definition statement and the include
statement which produces the output from the corresponding class.

The classes in this language support optional input parameters.

In contrast with other tools, the class is *not* a singleton, although
it can be used as one. Using include with equivalent input parameters
will cause the class to act as a singleton, although it can also be used
to produce distinct output.

The output produced by including a class is actually a list of
statements (a prog) which is ultimately a list of resources and edges.
This is different from functions which produces values.
2018-06-17 17:29:44 -04:00
James Shubin
83dab30ecf lang: Simplify bind stmt collection in the prog stmt
This cleans up the code to be more consistent with the other
improvements in this area.
2018-06-12 17:44:42 -04:00
James Shubin
24b08a332d pgraph: Handle empty graphs when merging two
In case we choose to add an empty (nil) graph, handle it safely. This
could allow us to return nil in a lang/structs Graph method without
issue.
2018-06-12 17:44:36 -04:00
James Shubin
70ccb3022a lang: Simplify struct interpolation
Cleaner code, nothing fancy.
2018-06-12 17:40:57 -04:00
James Shubin
8019b90b8a lang: Don't add identical resources to graph
This means that it's legal to produce two compatible (usually identical)
resources without a compile error and without causing two of them to get
run. It's too bad puppet never got this right.

It's probably worth checking if this could be done for edges too, and if
the logic can be contained in the engine and not in the frontend.
2018-06-12 17:40:57 -04:00
James Shubin
5f12ff6178 lang: Add indentation test to parser
This adds a test case to catch some common typos.
2018-06-12 17:40:18 -04:00
James Shubin
6e20e48489 lang: Simplify graph function for edge half in parser 2018-06-12 17:40:18 -04:00
James Shubin
f29a72235c lang: funcs: Registered functions map should be private
Make the map is private so that the public methods must be used to
access it.
2018-06-12 17:40:18 -04:00
James Shubin
e25d499eeb lang: Add edges to StmtProg output
I think I forgot to add these previously, and I think they should be
part of the output now.
2018-06-12 17:40:18 -04:00
James Shubin
9cae339546 lang: Error parser if SetType fails to avoid a panic
Turns out we can actually cause the parser to error instead of needing
to panic. It definitely seems to work, and is better than the panic. The
only awkward thing is how this plumbing works in yacc world. If anyone
knows why this is wrong, please let me know. Reading the generated code
seems to imply that this is correct.
2018-05-22 20:02:50 -04:00
James Shubin
a049af6262 engine: resources: print: Add missing Recvable trait
We we're receiving values, but we forgot to list the trait. This caused
an intentional engine panic, but is easily fixed :)
2018-05-22 19:32:40 -04:00
Jonathan Gold
a402f50f9b docs: Update url for AWS EC2 blog post 2018-05-19 22:05:12 -04:00
Jonathan Gold
9f89ea9be6 docs: Add netlink post to on-the-web.md 2018-05-19 22:05:12 -04:00
phaer
e538aacf9d vagrant: Fix example path in motd 2018-05-19 09:21:14 +02:00
phaer
968c609697 vagrant: Add gem package 2018-05-19 09:21:06 +02:00
phaer
c11cfa0a62 vagrant: Bump to fedora 28 2018-05-19 09:20:51 +02:00
Jonathan Gold
074f4677d5 build: Fix ldflags pattern for 1.10
Prior to go 1.10 ldflags would apply to all packages by default.
As of go 1.10 it is necessary to specify the package for the
flags to apply. This patch checks the go version, and formats
the build command accordingly.
2018-05-11 16:17:24 -04:00
James Shubin
9ea5c03371 travis: Enable apt updates on builds
This used to happen by default, and travis changed the default.
2018-05-09 13:46:04 -04:00
James Shubin
22c0ff3cf5 test: Improve golang tests with root and disabling cache
This allows golang tests to be marked as root or !root using build tags.
The matching tests are then run as expected using our test runner.

This also disables test caching which is unfriendly to repeated test
running and is an absurd golang default to add.

Lastly this hooks up the testing verbose flag to tests that accept a
debug variable.

These tests aren't enabled on travis yet because of how it installs
golang.
2018-05-09 13:44:01 -04:00
James Shubin
3ced981d28 engine: test: Pass in the go test verbose flag
This hooks up our debug variable to the go test verbose flag.
2018-05-09 12:11:35 -04:00
Jonathan Gold
299080f590 engine: DBus cleanup 2018-05-07 15:57:17 -04:00
James Shubin
a407771eaf test: Catch naked returns and check for canonically named imports
This catches scenarios where we forgot to prefix the error with return.
One of our contributors occasionally made this typo, and since core go
vet didn't (surprisingly) catch it, we should add a test!

It also adds a simple check for import naming aliases. Expanding this
test to add other cases and check for differently named values might
make sense.
2018-05-06 15:18:46 -04:00
Jonathan Gold
d26a6de759 engine: resources: mount: Add a mount resource 2018-05-04 15:53:05 -04:00
Jonathan Gold
9baad56197 util: Move dbus AddMatch const to util package 2018-05-04 15:46:14 -04:00
James Shubin
a589e2ecf3 docs, test: Remove old reference to resources package
Forgot to change this previously. Also updated the resources list in the
documentation.
2018-05-02 15:28:15 -04:00
Jonathan Gold
d7029871b1 engine: resources: nspawn: Remove godbus channel buffer
https://github.com/godbus/dbus/issues/94 is fixed with
https://github.com/godbus/dbus/pull/105, so the
buffered channel is no longer necessary.
2018-05-01 12:19:34 -04:00
Alan Jenkins
b80a505be5 engine: resources: packagekit: Add Arch mapping 'any' for Arch Linux compatibility
Arch Linux uses the mapping architecture name 'any'. This mapping was
missing from mgmt resulting in an error stating that arch 'any' did not
exist. Adding this mapping allows successful installation of packages
under Arch Linux.
2018-04-30 07:28:58 +01:00
James Shubin
412a25462e test: Improve commit message test
We can classify better now that we have the new engine.
2018-04-21 19:29:26 -04:00
James Shubin
9a8408a092 engine: Small fixes 2018-04-20 21:11:32 -04:00
James Shubin
86a9181e9b puppet: Clean up the GAPI and remove log package
This uses the proper facilities which makes things a bit more uniform.
2018-04-19 01:56:31 -04:00
James Shubin
9969286224 engine: Resources package rewrite
This giant patch makes some much needed improvements to the code base.

* The engine has been rewritten and lives within engine/graph/
* All of the common interfaces and code now live in engine/
* All of the resources are in one package called engine/resources/
* The Res API can use different "traits" from engine/traits/
* The Res API has been simplified to hide many of the old internals
* The Watch & Process loops were previously inverted, but is now fixed
* The likelihood of package cycles has been reduced drastically
* And much, much more...

Unfortunately, some code had to be temporarily removed. The remote code
had to be taken out, as did the prometheus code. We hope to have these
back in new forms as soon as possible.
2018-04-19 01:10:58 -04:00
James Shubin
ef49aa7e08 lang: Don't race with a ^C to the obj.lang calls
If we trigger a close, we must not run the LangClose before we've exited
from the loop, because that loop could race and run code which depends
on LangClose not having run first. So run the loop shutdown, then let
the wait group expire, before shutting down the lang.
2018-04-16 08:38:22 -04:00
James Shubin
acdb497b80 etcd: Pull in default URLs from upstream
This depends on https://github.com/coreos/etcd/pull/6837
2018-04-16 08:38:22 -04:00
James Shubin
4d8faeb826 lib, yamlgraph: Remove old yamlgraph GAPI frontend
I should have removed this a long time ago, but didn't. Now it's done.
The new v2 frontend is loosing the v2 name and just replacing v1.
2018-04-16 08:38:22 -04:00
James Shubin
6e0dfdb16f lib: Remove hcl GAPI frontend
This is currently unmaintained and the normal mcl language exists which
is preferable to this. As a result, I'm removing this for now to make an
upcoming refactor easier. We can add it back easily if someone has
interest.
2018-04-16 08:38:22 -04:00
James Shubin
754480a9b6 readme: Add patreon link to README file 2018-04-16 08:37:49 -04:00
jesus m. rodriguez
15681ddca9 build: Add help to main Makefile 2018-04-08 23:09:47 -04:00
Jonathan Gold
3c8d424a43 util: Rename SortedStrSliceCompare and move to util package 2018-03-29 00:55:18 -04:00
jonathangold
7d7eb3d1cd resources: net: Add net resource
This patch adds a net resource for managing nework interfaces, based
around netlink.
2018-03-27 17:46:00 -04:00
James Shubin
8500339ba6 lang: Add mutex around Expr String/Value/SetValue calls
The golang race detector complains about some unimportant races, and as
a result, this patch adds some mutexes to prevent these test failures.
We actually lock more than necessary, because a more accurate version
would be more time consuming to implement. Secondarily, it's likely that
in the future we replace this function graph algorithm with something
that is guaranteed to be glitch-free and supports back pressure.
2018-03-27 15:30:59 -04:00
James Shubin
06ee05026b lang: funcs: Don't race when building an initial graph
I noticed a very intermittent test failure where interpret would end up
running, but *fail* because a value wasn't present. This should never
happen, because the function engine is designed to only call interpret
when there has been at least one value produced for every node in the
AST. So what is the bug that would produce:

interpret error: could not interpret: func value does not yet exist

About 20 minutes ago while I was getting to bed, it occurred to me where
to look! Out of bed and to the laptop, and after briefly reminding
myself of the code, I think I've found the issue.

What I think was happening, was that an AST node would produce a value,
and send a message on the aggregate channel. This channel is monitored,
and every time it receives a message, it checks to ensure that all the
values now exist before producing a message for interpret to run.
However, this AST node was not the final one to be produced, but before
the message was read by the aggregate channel, the last remaining AST
node ran and set it's "loaded" state to `true`, but *before* its value
was made available for the aggregate channel to read. That channel then
occasionally won the race and tried to access a value before it existed,
thus causing out intermittent bug.

At least I think that's what was going on. Hopefully this patch fixes
this, if not, then there's another bug hiding too! And of course, this
entire function engine could do with some proper analysis from someone
familiar with glitches, back pressure, and FRP parallelism.

One particular note was that I used my brain, not some fancy debugging
tool to find this. Maybe skilled debuggers can fork lift their tools
onto this type of problem, but I haven't those skills!

¯\_(ツ)_/¯
2018-03-15 23:22:21 -04:00
James Shubin
ddefb4e987 integration: Log the instance output
This adds logging so that you can dig deeper into crashes or issues.
2018-03-13 06:38:21 -04:00
James Shubin
62d1fc7ed3 test, integration: Add cluster primitives to integration framework
This further extends the integration framework to add some simple
primitives for building clusters. More complex primitives and patterns
can be added in the future, but this should serve the general cases.
2018-03-13 06:38:21 -04:00
James Shubin
f3b99b3940 test, integration: Add an integration test framework
This adds an initial implementation of an integration test framework for
writing more complicated tests. In particular this also makes some small
additions to the mgmt core so that testing is easier.
2018-03-13 06:38:21 -04:00
Lauri Ojansivu
97c11c18d0 resources: svc: Add activating state
There seems to be a "activating" state that some services can reach.
Related #369
2018-03-10 15:27:07 +02:00
James Shubin
93a909551f recwatch: Remove the ConfigWatch functionality
This is some now dead code which was buggy and badly written. Time to
get rid of unnecessary technical debt so that we can move forward!
2018-03-09 22:26:10 -05:00
James Shubin
ea52eb78d9 lib: Remove remote execution from core
I have an improved design for remote execution as a resource. Since I
need to get rid of some technical debt to clean up the resource API, and
this main loop, a good first step is to remote it's invocation. It will
be coming back as a resource as soon as possible!
2018-03-09 17:07:58 -05:00
James Shubin
fdd698dade resources: svc: Add deactivating state
There seems to be a "deactivating" state that some services can reach.
Add this case, and switch the panic to an error.
2018-03-09 17:04:30 -05:00
James Shubin
173ccf6861 pgraph: Don't panic on new or nil graphs
This adds a bit of flexibility so that we can still run a topological
sort on a nil graph.
2018-03-05 01:58:43 -05:00
James Shubin
a5c3db6303 lang: Misc fixes for typos and grammar 2018-02-28 00:35:22 -05:00
James Shubin
3ad7097c8a lang: Add internal, resource specific edges
This adds the ability to specify internal, resource specific edges, with
and without notifications. We use the special words: "Notify", "Before",
"Listen", and "Depend". They must have the first character capitalized.
They also support the "elvis" operator.
2018-02-27 23:26:25 -05:00
James Shubin
8e01b6db48 lang: Add a resource-specific elvis operator
This allows you to omit a resource parameter programmatically, and
avoids the need of an `undef` or `nil` in our language, which would
contribute to programming errors, crashes, and overall reduced safety.
2018-02-27 17:29:49 -05:00
James Shubin
67607eba8b travis: Fix the OSX builds
I don't use OSX, but here's a bit of sympathy for the poor travis OSX
builder that can't understand apt ;)
2018-02-27 17:28:09 -05:00
James Shubin
6e7a71d01a travis: Attempt to cut down on flaky failures
Travis has been spuriously failing a LOT. Hopefully this reduces some of
those failures.
2018-02-27 17:17:29 -05:00
James Shubin
ff69a82b57 lang: unification: Fix panic in struct/func cmp of partials
This was discovered by user aequitas. I modified his patch slightly, and
added some comments and a test.
2018-02-27 16:39:10 -05:00
James Shubin
df1e50e599 lang: funcs: Add math pow function and a few examples
Just a few small things I think should be committed.
2018-02-25 19:48:25 -05:00
James Shubin
6370f0cb95 lang: Add edges to lexer and parser
This adds some initial syntax for external edges to the language.

There are still improvements which are necessary for send/recv.
2018-02-25 19:29:27 -05:00
James Shubin
80784bb8f1 lang: types, funcs: Add simple polymorphic function API
This adds a simple API for adding static, polymorphic, pure functions.
This lets you define a list of type signatures and the associated
implementations to overload a particular function name. The internals of
this API then do all of the hard work of matching the available
signatures to what statically type checks, and then calling the
appropriate implementation.

While this seems as if this would only work for function polymorphism
with a finite number of possible types, while this is mostly true, it
also allows you to add the `variant` "wildcard" type into your
signatures which will allow you to match a wider set of signatures.

A canonical use case for this is the len function which can determine
the length of both lists and maps with any contained type. (Either the
type of the list elements, or the types of the map keys and values.)

When using this functionality, you must be careful to ensure that there
is only a single mapping from possible type to signature so that the
"dynamic dispatch" of the function is unique.

It is worth noting that this API won't cover functions which support an
arbitrary number of input arguments. The well-known case of this,
printf, is implemented with the more general function API which is more
complicated.

This patch also adds some necessary library improvements for comparing
types to partial types, and to types containing variants.

Lastly, this fixes a bug in the `NewType` parser which parsed certain
complex function types wrong.
2018-02-25 02:17:13 -05:00
James Shubin
40dcd6ec99 all: Misc fixes and test fixes 2018-02-25 02:13:51 -05:00
karimb
06f2d65500 docs: Add docs for docker usage 2018-02-24 12:04:23 +01:00
Johan Bloemberg
98d3c299ff project: Add me 2018-02-23 19:59:55 -05:00
James Shubin
46da3a34a0 docs: Add two new faq entries 2018-02-23 19:44:54 -05:00
Johan Bloemberg
f33f84d2f2 lang: Add getenv function
$x = getenv("NAME")
    $y = defaultenv("NOTEXIST", "defaultvalue")
    $z = hasenv("NAME")
    $a = env()
    $b = maplookup($a, "NAME", "defaultvalue")
2018-02-23 20:02:13 +01:00
James Shubin
a785a43ef3 travis: Attempt to workaround the constant travis failures
I'm beginning to think we need a more reliable CI...
2018-02-22 20:26:52 -05:00
James Shubin
b0911c6d70 lang: funcs: simple: Don't block on simple, pure, static functions
I forgot to handle the special case of a function using this API that
received no inputs. It was waiting for the first input to come in, and
as a result was never producing any output.

Remember that functions like this should *almost* be thought of as
constants of the system. You would expect their output to never change
during the lifetime of a particular program invocation.
2018-02-22 19:26:18 -05:00
James Shubin
81a0e9e8c7 build: Relocate time command to the front for readability
This makes the output more readable in my terminal.
2018-02-22 17:49:33 -05:00
Johan Bloemberg
06d33a45f5 docs, misc: Add tool references, .editorconfig for mcl 2018-02-22 17:23:11 -05:00
Jonathan Gold
cfb8deac56 project: Add Jonathan Gold to AUTHORS 2018-02-22 17:19:58 -05:00
Johan Bloemberg
9544ab2e02 recwatch: Fix watching newly created files on macOS
Fixes: https://github.com/purpleidea/mgmt/issues/33
2018-02-22 16:52:26 -05:00
Oliver Frommel
318fe4a5dc misc: Small fixes for makedeps script
- install Go distribution package only if no go binary found
2018-02-22 16:50:13 -05:00
James Shubin
5597183391 docs: Add two faq entries about the type system 2018-02-22 16:45:54 -05:00
James Shubin
05c60d9a59 test, docs: Restrict long lines in markdown linter
It's getting out of hand...
2018-02-22 16:19:23 -05:00
Peter Oliver
f01eea33e9 emacs: Bundle an Emacs major mode, mgmtconfig-mode
This provides syntax highlighting, commenting, and rudimentary indentation of the mgmt language.
2018-02-22 15:57:05 -05:00
James Shubin
9992c367bf misc: Update golint to new location
Somehow this got changed...
2018-02-22 02:01:14 -05:00
James Shubin
d275a23a81 misc: Add dependency on time package
Some environments apparently don't have this installed. We have it in
certain places where we like to time things.
2018-02-21 22:52:41 -05:00
James Shubin
14ddd7c196 golint: Fix ineffassign mistakes 2018-02-21 22:52:41 -05:00
James Shubin
0815b20b76 lang: funcs: Fix up some old comments
Woops, bad copy-paste issues.
2018-02-21 22:52:41 -05:00
James Shubin
cffdb06181 test, docs: Add a linter for testing markdown, and fix up our docs
While writing docs, I couldn't remember what the correct style was
supposed to be, and I remember someone complaining about this
previously, so I decided to add a linter! I excluded a bunch of annoying
style rules, but if we find more we can add those to the list too.

Hopefully this gives us a more consistent feel throughout.
2018-02-21 22:52:41 -05:00
James Shubin
837388ae4e lang: types, funcs: Add simple function API
This patch adds a simple function API for writing simple, pure
functions. This should reduce the amount of boilerplate required for
most functions, and make growing a stdlib significantly easier. If you
need to build more complex, event-generating functions, or statically
polymorphic functions, then you'll still need to use the normal API for
now.

This also makes all of these pure functions available automatically
within templates. It might make sense to group these functions into
packages to make their logical organization easier, but this is a good
enough start for now.

Lastly, this added some missing pieces to our types library. You can now
use `ValueOf` to convert from a `reflect.Value` to the corresponding
`Value` in our type system, if an equivalent exists.

Unfortunately, we're severely lacking in tests for these new types
library additions, but look forward to growing some in the future!
2018-02-21 21:32:31 -05:00
Johan Bloemberg
cbd2bdd4c5 travis: Retry flaky apt update at build start 2018-02-20 21:41:08 +01:00
Johan Bloemberg
f34ca3a5ca travis: Improve travis speed by only building 1 go version for osx 2018-02-20 21:41:03 +01:00
James Shubin
4898297cce travis: Avoid notification noise from forks
Encrypt name of IRC channel to workaround forks spamming us with their
testing messages.

Docs: https://docs.travis-ci.com/user/environment-variables/#Defining-encrypted-variables-in-.travis.yml
2018-02-20 14:12:09 -05:00
Johan Bloemberg
ffcc2aa2af lib: Provide detailed feedback about invalid URLs 2018-02-20 10:29:19 -05:00
Johan Bloemberg
158fb8d31c etcd: Warn about invalid configuration, clarify --no-server 2018-02-20 10:29:19 -05:00
Johan Bloemberg
07714c67cb cli: Log errors return by Run functions
Turns

```
$ ./mgmt run
00:44:15 hello.go:46: This is: mgmt, version: 0.0.14-30-ge3a2648
00:44:15 hello.go:47: Main: Start: 1518738255855525279
$
```

Into

```
$ ./mgmt run
01:07:02 hello.go:46: This is: mgmt, version: 0.0.14-30-ge3a2648-dirty
01:07:02 hello.go:47: Main: Start: 1518739622517652739
01:07:02 cli.go:167: Main: Error: can't create prefix: mkdir /var/lib/mgmt/: permission denied
$
```
2018-02-20 10:29:19 -05:00
James Shubin
f12e502c61 lang: funcs: Rename things for consistency
Also fix a few copy-pasta issues in the documentation.
2018-02-18 19:47:14 -05:00
Toshaan Bharvani
2fdf8d5dc3 lang: Interface sorting order
as golang does not loop over the same map/list always the same
we use a helper list to sort it

Signed-off-by: Toshaan Bharvani <toshaan@vantosh.com>
2018-02-18 18:32:15 -05:00
James Shubin
28ec7a1e54 etcd: scheduler: Remove etcd 3.2 specific hacks
Now that we're using etcd 3.3, we can simplify our code now that our
patches are in a release.
2018-02-18 18:28:45 -05:00
James Shubin
24cb2e6450 etcd: Increase the default max txn op count
The default of 128 is fairly low for large code bases. Please let us
know if you hit the new limit of 512.
2018-02-18 18:28:00 -05:00
James Shubin
915b022901 test: Show test output as it happens 2018-02-18 17:31:45 -05:00
James Shubin
4a623c1891 docs: Add an entry to the faq about converged timeouts 2018-02-18 16:28:50 -05:00
Johan Bloemberg
7508161c39 test: Exclude generated files from golint 2018-02-18 15:07:54 -05:00
Johan Bloemberg
d33861ccb4 test: Fix augeas test for macOS, improve test debuggability
- resolve a discrepancy in augeas behaviour on macOS
- on macOS `sed` requires an argument for `-i`.
- made the test fail as early as it can
- provide information about why the test is failing
2018-02-18 14:36:59 -05:00
Johan Bloemberg
572b2575c5 test: Export the mgmt command to be used during test 2018-02-18 14:19:05 -05:00
James Shubin
d99190b166 travis: Add golang 1.10.x to builds 2018-02-18 13:58:11 -05:00
James Shubin
bc91b03276 docs: Add faq entry about production readiness 2018-02-18 13:34:21 -05:00
James Shubin
9ba893c06c etcd: Bump to etcd v3.3 and golang 1.9
This moves us to etcd v3.3 (a new major release) which has some useful
features but that requires version 1.9 of golang.
2018-02-15 18:47:55 -05:00
James Shubin
27e51f1bcb authors: Clarify wording in AUTHORS file 2018-02-15 18:45:32 -05:00
James Shubin
e3a26483e8 test: Improve gometalinter test so that it skips generated files
This should improve things significantly, and avoid the failures now
that we're testing after the files have already been built.
2018-02-15 16:26:06 -05:00
James Shubin
33a4fd6fbe build: Add -i flag to go build
It got accidentally dropped, but is crucial for happiness.
See: https://purpleidea.com/blog/2017/02/26/faster-golang-builds/
2018-02-15 12:24:56 -05:00
Johan Bloemberg
b34b359860 test: Streamline test suite a little
This change aims to streamline the integrationtest suite and reduce friction when running (parts of) test suites.

Changes:
- add `test-testname` to makefile to easily run one suite
- made skipping tests first class citizen in test.sh (all available testsuites and the reasons they are skipped are now better exposed and discovered)
- suppress some output of gotest unless there is an error
- no longer build binary for examples and gotest suites
- removed .SILENT from makefile as it being applied to only some targets makes it feel weird (I just learned about this option btw, feel free to comment on this change)
- move individual tests out of `test.sh` and into `test-misc.sh`
- introduced the concept of testsuites to `test.sh`
2018-02-15 17:21:49 +01:00
James Shubin
8b9491823d etcd: Fix golint issue in test
Found with new gometalinter version.
2018-02-14 18:53:10 -05:00
James Shubin
b8b6e5266f build: Improve speed of make
Generating a huge amount of unnecessary targets caused "noop" make runs
to take seven seconds on my machine. This limits the list of these
drastically and now "noop" make's are now < 1s on my machine.

Issue discussed in:
https://github.com/purpleidea/mgmt/issues/331
2018-02-14 18:42:12 -05:00
James Shubin
5f80c1ac2a resources: nspawn: Don't panic if one svc is nil
Not sure why one of them was nil, but this prevents the panic.
2018-02-14 16:44:21 -05:00
James Shubin
3af7e815d0 docs: Add newly recorded talks and blog posts about mgmt 2018-02-14 15:45:40 -05:00
James Shubin
714afe35a1 test: Fix broken gometalinter test
The test for gometalinter got silently broken in an earlier commit.
Look for the missing space that was added back in this commit to see
why! In any case, this now fixes some of the things that weren't
previously caught by this change.

If anyone knows how to run these sorts of tests properly so that entire
packages are tested and so that we can enable additional tests, please
let me know!

It's also unclear why goreportcard catches a few additional problems
which aren't found by running this ourselves.

See:
https://goreportcard.com/report/github.com/purpleidea/mgmt
for more information.
2018-02-14 14:34:36 -05:00
James Shubin
b0a8f585c3 readme: Fix broken link 2018-02-14 14:03:07 -05:00
James Shubin
1a2918082d docs: Add FAQ entry about vendoring dependencies 2018-02-14 14:01:10 -05:00
Johan Bloemberg
22e4dfa534 build: Unify build/crossbuild
Changes:

- allows explicit crossbuild targets (eg: `make mgmt-darwin-amd64`)
- adds darwin/amd64 to default crossbuild targets
- gitignore only build artifacts (eg: not all files starting with `mgmt-`)
- `build` and `crossbuild` target now utilize the same build function (`build` still generates only a `mgmt` binary for the current os/arch)
- test crossbuilding
- allow specifying custom GOOSARCHES envvar to override defaults
- crossbuild artifacts go into `build/` now
- add `build-debug` which includes symbol tables and debug info
- the build function now has `-s -w` linker arguments which discards some debug info afaict, to build a debug release use `make build-debug`

On my mac crossbuilding won't work unless I disable augeas and libvirt:

```
~/.g/s/g/p/mgmt (build|●1✚8…3) $ make build
Generating: bindata...
Generating: lang...
/Applications/Xcode.app/Contents/Developer/usr/bin/make --quiet -C lang
Building: mgmt, os/arch: darwin-amd64, version: 0.0.14-12-g94c8bc1-dirty...
env GOOS=darwin GOARCH=amd64 time go build -ldflags "-X main.program=mgmt -X main.version=0.0.14-12-g94c8bc1-dirty -s -w" -o mgmt-darwin-amd64 ;
        7.14 real        10.36 user         1.73 sys
mv mgmt-darwin-amd64 mgmt
```

```
~/.g/s/g/p/mgmt (build|●1✚8…3) $ time env GOTAGS='noaugeas novirt' make crossbuild
Generating: bindata...
Generating: lang...
/Applications/Xcode.app/Contents/Developer/usr/bin/make --quiet -C lang
Building: mgmt, os/arch: linux-amd64, version: 0.0.14-12-g94c8bc1-dirty...
env GOOS=linux GOARCH=amd64 time go build -ldflags "-X main.program=mgmt -X main.version=0.0.14-12-g94c8bc1-dirty -s -w" -o mgmt-linux-amd64 -tags 'noaugeas novirt';
       18.48 real        50.02 user         5.83 sys
Building: mgmt, os/arch: linux-ppc64, version: 0.0.14-12-g94c8bc1-dirty...
env GOOS=linux GOARCH=ppc64 time go build -ldflags "-X main.program=mgmt -X main.version=0.0.14-12-g94c8bc1-dirty -s -w" -o mgmt-linux-ppc64 -tags 'noaugeas novirt';
       29.83 real        85.09 user        11.54 sys
Building: mgmt, os/arch: linux-ppc64le, version: 0.0.14-12-g94c8bc1-dirty...
env GOOS=linux GOARCH=ppc64le time go build -ldflags "-X main.program=mgmt -X main.version=0.0.14-12-g94c8bc1-dirty -s -w" -o mgmt-linux-ppc64le -tags 'noaugeas novirt';
       29.74 real        85.84 user        11.76 sys
Building: mgmt, os/arch: linux-arm64, version: 0.0.14-12-g94c8bc1-dirty...
env GOOS=linux GOARCH=arm64 time go build -ldflags "-X main.program=mgmt -X main.version=0.0.14-12-g94c8bc1-dirty -s -w" -o mgmt-linux-arm64 -tags 'noaugeas novirt';
       28.33 real        83.24 user        11.40 sys
Building: mgmt, os/arch: darwin-amd64, version: 0.0.14-12-g94c8bc1-dirty...
env GOOS=darwin GOARCH=amd64 time go build -ldflags "-X main.program=mgmt -X main.version=0.0.14-12-g94c8bc1-dirty -s -w" -o mgmt-darwin-amd64 -tags 'noaugeas novirt';
        7.16 real        10.15 user         1.74 sys
      114.71 real       315.26 user        42.44 sys
```
2018-02-14 12:37:49 -05:00
Johan Bloemberg
41eb850b3d debian: Add graphviz and packagekit runtime dependencies 2018-02-12 15:11:13 -05:00
Wim
3a50171d19 build: Add gcc,pkg-config deps 2018-02-12 15:10:19 -05:00
Wim
6c9e0ff974 docs: Add GOPATH/bin to PATH 2018-02-12 15:09:32 -05:00
James Shubin
644e5164b1 test: Increase timeouts for when travis is slow
Hopefully this cuts down on spurious failures.
2018-02-12 15:08:47 -05:00
James Shubin
4fefa9f2f0 travis: Disable fast finish for now
This causes a notification for each entry in the matrix which is now too
many emails. When travis adds and option to send just one notification,
but to still allow you to fast finish, then please lmk :)
2018-02-10 18:49:12 -05:00
Johan Bloemberg
4c793e0ee6 misc: Fix graphviz output for hostnames with dot in them 2018-02-11 00:02:52 +01:00
James Shubin
68a7de41ae etcd: Update broken link 2018-02-10 10:38:54 -05:00
dsx
94c8bc1de9 debian: Add packaging 2018-02-10 05:12:31 -05:00
Johan Bloemberg
8fb0373f82 resources: Do not return GID for UID lookup
On linux it is convention for users to have a group with the same GID as the users UID. On macOS this is not the case. This broke the test which lead to discovering this bug.
2018-02-10 05:01:12 -05:00
Johan Bloemberg
d567dc3769 lang: Use universal way to retrieve load
Sysinfo is not supported on macOS and results in a build error.
2018-02-10 05:01:12 -05:00
Johan Bloemberg
ba21554c5f build, docs: Improve macOS building
- New docker command for quickly running tasks in a Linux environment.
- Updated docs with macOS specific details.
- Fixed some test issues.
- Add (fallible) macOS test target for Travis.
2018-02-10 05:01:12 -05:00
James Shubin
e37bb3ac8a test: Add new test for language prefix 2018-02-07 21:52:06 -05:00
Carsten Thiel
79845f0dfd test: Refactor unification_test to subtests
Testsuite for unification now uses subtests feature.
2018-02-07 16:13:18 +01:00
Carsten Thiel
eb33a5a5df docs: Improve file resource documentation
Info on how to create a directory.
Explain more parameter options.
2018-02-07 14:19:14 +01:00
jonathangold
adbe9c7be1 misc: Replace missing go-bindata dependency 2018-02-07 06:15:03 -05:00
James Shubin
b19583e7d3 lang: Initial implementation of the mgmt language
This is an initial implementation of the mgmt language. It is a
declarative (immutable) functional, reactive, domain specific
programming language. It is intended to be a language that is:

* safe
* powerful
* easy to reason about

With these properties, we hope this language, and the mgmt engine will
allow you to model the real-time systems that you'd like to automate.

This also includes a number of other associated changes. Sorry for the
large size of this patch.
2018-01-20 08:09:29 -05:00
Joe Julian
1c8c0b2915 misc: Don't install packages that are already installed
pacman needs `--needed` to prevent reinstalling packages that are
already installed. Additionally I added `--asdeps` to allow later
cleanup of unneeded dependencies.
2018-02-03 17:17:23 -05:00
Joe Julian
bee1aa00f1 misc: Use bash's command instead of which
Bash has a built-in command, `command`, that will search the path and
return the full path to a command if it exists (or an exit code of 1 if
it does not), preventing the requirement of the `which` package.
2018-02-03 11:23:52 -05:00
Toshaan Bharvani
077b6e540a build: Add cross building option
Added a cross build option using a buildrelease function

Signed-off-by: Toshaan Bharvani <toshaan@vantosh.com>

build: Add gitignore entry for mgmt-* binaries

Signed-off-by: Toshaan Bharvani <toshaan@vantosh.com>

build: Update makefile based upon feed back

* rename cross to crossbuild
* added crossbuild to PHONY

Signed-off-by: Toshaan Bharvani <toshaan@vantosh.com>

build: Change the order of .PHONY as per the rest of the file

Signed-off-by: Toshaan Bharvani <toshaan@vantosh.com>
2018-02-03 16:43:40 +01:00
James Shubin
e8b03545bb test: Don't fail on tag builds
This seems to be causing our failures with:

$ git fetch --unshallow
fatal: Couldn't find remote ref refs/heads/0.0.x

where x is some tag.

Hopefully this doesn't break the other use case we added this patch for!
2018-01-11 18:03:56 -05:00
James Shubin
70c59eab4a misc: Don't display script name in output 2018-01-11 18:03:04 -05:00
jonathangold
3c677543e0 resources: aws: ec2: Fix closed channel handling
If awschan closes, longpollWatch and snsWatch return nil
instead of an error. This will prevent the engine from
shutting down in case we choose to close the channel
early or from other struct methods.
2018-01-06 15:15:30 -05:00
jonathangold
c455ef2c62 resources: aws: ec2: Send IP addresses and InstanceID 2018-01-03 21:34:28 -05:00
Jonathan Gold
032d0992d6 resources: aws: ec2: Refactor CheckApply
CheckApply was rewritten, using the new describe methods to improve
readability and maintainability.
2018-01-03 21:34:28 -05:00
jonathangold
67837a47ac resources: aws: ec2: Refactor longpollWatch
Complete rewrite of longpollWatch() for correctness and maintanability.
2018-01-03 21:34:28 -05:00
Jonathan Gold
32e3c4e029 resources: aws: ec2: Refactor longpollWatch
This patch simplifies longpollwatch by getting rid of some unnecessary
api calls and breaking the waiters out into their own functions.
2018-01-03 21:34:28 -05:00
Jonathan Gold
76fcb7a06e resources: aws: ec2: Wait for stop and terminate concurrently
In longpollWatch it was no longer sufficient to use only
WaitUntilInstanceStopped as it would block if the instance was
terminated. This patch launches two goroutines in its place, one
waits until the instance stops and the other waits until it
terminates. When either one returns, it cancels their context,
and execution continues.
2018-01-03 21:34:28 -05:00
Jonathan Gold
149a2188e2 resources: aws: ec2: Retry on exceeded wait attempts error
The waiters now return the AwsErr error "ResourceNotReady: exceeded wait
attempts" when the instance state does not converge after 40 retries.
During longpollWatch() we need to detect this error and continue to
the top of the loop so we can restart the waiters and keep watching for
events.
2018-01-03 21:34:28 -05:00
Jonathan Gold
08e7caea6b resources: aws: ec2: CheckApply fix pending and stopping cases
If CheckApply was called when the instance was pending or stopping, it
would return an error. This patch supresses these errors and tells the
engine that the state can't yet be changed.
2018-01-03 21:34:28 -05:00
Jonathan Gold
e330ebc8c9 resources: aws: ec2: Verify SNS message signatures 2018-01-03 21:34:28 -05:00
Jonathan Gold
388a08e13a resources: aws: ec2: Check that policy.Statement != nil 2018-01-03 21:34:28 -05:00
Jonathan Gold
9ba9ef1cbf resources: aws: ec2: Close closeChan before server shutdown
This patch makes sure that closeChan is closed as soon as the main loop
returns, so any channel operations are unblocked before we run shutdown.
This ensures that the server's goroutine can return before shutdown
completes and we don't panic by trying to serve the client after
shutdown returns.
2018-01-03 21:34:27 -05:00
Jonathan Gold
fac004b774 resources: aws: ec2: Update postHandler to process messages 2018-01-03 21:34:27 -05:00
Jonathan Gold
8cd3f28734 resources: aws: ec2: Authorize CloudWatch to publish to sns 2018-01-03 21:34:27 -05:00
Jonathan Gold
dcd23fcf75 resources: aws: ec2: Add CloudWatch rule and target SNS
This patch creates the cloudwatch rule that detects ec2 instance
state changes, and targets the rule to publish on our sns topic
which, in turn, pushes those event notifications to our endpoint.
2018-01-03 21:34:27 -05:00
Jonathan Gold
1162485c2c resources: aws: ec2: Subscribe SNS endpoint to topic
This patch adds methods to subscribe and confirm the subscription
to the sns topic.
2018-01-03 21:34:27 -05:00
Jonathan Gold
966172eac6 resources: aws: ec2: Use custom listener for snsServer
This patch replaces the call to Server.ListenAndServe() with
Server.Serve(listener) in order to make sure the listener is up
and running before we subscribe to the topic in a future patch.
2018-01-03 21:34:27 -05:00
James Shubin
12fce52cd7 legal: Happy 2018 everyone...
Done with:

ack '2017+' -l | xargs sed -i -e 's/2017+/2018+/g'

Checked manually with:

git add -p

Hello to future James from 2019, and Happy Hacking!
2018-01-03 21:22:07 -05:00
Felix Frank
5ca1e2a23f puppet: Avoid empty parameters to puppet mgmtgraph
This solves an issue first observed with golang 1.8.

Creating an exec.Command with an empty string parameter (when no puppet.conf
file is specified) would lead to an error from Puppet, stating that an
unexpected argument was passed to "puppet mgmtgraph print".

The workaround is to not include *any* positional argument (not even the
empty string) when --puppet-conf is not used.
2017-12-26 00:18:46 +01:00
Paul Morgan
98f8a61e83 git: Configure editorconfig to indent with tabs in bash scripts
This follows `test/test-bashfmt.sh` style check(s).
2017-12-20 21:09:15 +00:00
Paul Morgan
2e86d7c5ab git: Ensure the tagging script is idempotent 2017-12-20 21:04:57 +00:00
Jonathan Gold
62ca12608d cli: Add license flag
This patch adds the option to print the license with a cli flag. It
uses go-bindata to store the license file. The file is generated by
running `make bindata` and the result is stored in the bindata
directory.
2017-12-08 00:57:58 -05:00
Jonathan Gold
406aa55667 resources: virt: Update libvirt-xml target
Builds started failing due to go-libvirt-xml 6d97448. In that patch,
the DomainChannelTarget struct was changed from having a single type
field, to having an individual field for each virtualization type.

This patch updates the connection check in Init to reflect the changes
to go-libvirt-xml, so that builds no longer fail.
2017-11-29 19:03:56 -05:00
James Shubin
a76dce8b15 docs: Add missing blog post about augeas resource 2017-11-26 17:15:49 -05:00
James Shubin
b01d453ae3 docs: Refresh documentation to provide a better new user experience
This does some cleanups and moves some things around for a better
experience. If you're an expert in this area, or are a new user who has
some feedback about their first impressions and experiences, please let
us know!
2017-11-25 20:45:57 -05:00
Guillaume Herail
ac629404f4 test: Switch to goimports instead of gofmt
see https://github.com/purpleidea/mgmt/pull/256#issuecomment-346360414
2017-11-25 06:49:00 -05:00
Guillaume Herail
3575d597f7 resources: Add User/Group to ExecRes 2017-11-24 10:38:16 -05:00
Toshaan Bharvani
2affcba3b4 build: Added build option to strip binary
This is a build option in Golang that will strip the binary.
The binary becomes about 50% smaller.

Signed-off-by: Toshaan Bharvani <toshaan@vantosh.com>
2017-11-24 10:26:48 -05:00
James Shubin
846c5f8762 test: Add another check for off-by-one-error commit tags 2017-11-24 09:46:32 -05:00
Julien Pivotto
086af712d2 example: Remove content out of directory definition
Signed-off-by: Julien Pivotto <roidelapluie@inuits.eu>
2017-11-24 14:26:20 +01:00
Julien Pivotto
2b6e39f283 build: Remove go 1.3 and 1.4 support
Signed-off-by: Julien Pivotto <roidelapluie@inuits.eu>
2017-11-24 05:35:09 -05:00
Julien Pivotto
472663193a prometheus: Initialize all metrics
Signed-off-by: Julien Pivotto <roidelapluie@inuits.eu>
2017-11-24 11:02:36 +01:00
James Shubin
879ff838ae resources: Replace golang 1.6 specific code with newer 1.7 version
We now require at least 1.8 so we might as well fix this up.
2017-11-23 10:57:11 -05:00
Julien Pivotto
5e9a085e39 exec: Add autoEdges between ExecRes and PkgRes
Signed-off-by: Julien Pivotto <roidelapluie@inuits.eu>
2017-11-23 16:30:22 +01:00
Julien Pivotto
c2b5729ebd build: Build mgmt on any go file change
Prior to this commit, running make would only rebuild mgmt when
main.go was changed. It means that make clean build was needed.

With this commit, any go file change in this directory will
trigger a new compilation.

Signed-off-by: Julien Pivotto <roidelapluie@inuits.eu>
2017-11-23 09:32:02 -05:00
Julien Pivotto
fdce9d6a6a prometheus: Initialize mgmt_checkapply_total metrics
It is recommended by Prometheus to initialize metrics:

https://prometheus.io/docs/practices/instrumentation/#avoid-missing-metrics

This commits initialize the mgmt_checkapply_total metric
for each registered resource.

Signed-off-by: Julien Pivotto <roidelapluie@inuits.eu>
2017-11-23 15:23:41 +01:00
Guillaume Herail
bfc2549289 resources: Move FileRes.uid()/.gid() to util.go 2017-11-23 08:34:38 -05:00
James Shubin
52fd1ae73e test: Add check for common doc vs docs ambiguity 2017-11-23 08:20:44 -05:00
Julien Pivotto
23e167616f doc: Fix link to the prometheus wiki
Signed-off-by: Julien Pivotto <roidelapluie@inuits.eu>
2017-11-23 09:52:28 +01:00
James Shubin
51ce83f20b test: Add extra commit message tests for some common mistakes
Feel free to add more if we identify them.
2017-11-21 11:05:20 -05:00
Jonathan Gold
5e5bbf4b39 travis: Allow travis builds to access target branches
Because travis builds only fetch a single branch (master) by default,
test-commit-message.sh only had access to commits in the master branch.
In order to fetch the correct branch for our build, we need to run
'git config remote.origin.fetch..' with the target branch's information
before executing git fetch on the repo in before_install.

Now git will always fetch the appropriate branch.
2017-11-18 21:04:12 -05:00
Guillaume Herail
cbc3a691b9 docker: Bump to golang 1.8 2017-11-16 17:36:35 +01:00
Jonathan Gold
a5247d6e69 resources: aws: ec2: Change event messages to iota consts 2017-11-14 16:48:51 -05:00
Jonathan Gold
d698b82a83 resources: aws: ec2: Start and stop SNS endpoint in snsWatch
This patch adds snsWatch which launches the HTTP server and listens
for messages on awsChan to forward as events to the mgmt engine.
2017-11-11 23:07:12 -05:00
Jonathan Gold
91eff75288 resources: aws: ec2: Add method to make sns topic 2017-11-10 17:31:19 -05:00
James Shubin
91a9edb322 resources: aws: ec2: Fix deadlock on rare error scenarios
If we get an error in the Watch loop, it will send this on awsChan,
which will cause Watch to loop. However, in this scenario it will never
cause closeChan to close, and we will deadlock because we have a
waitGroup in a helper goroutine which is waiting on this channel to
close the context.

Normally this wouldn't be an issue, but since we have more than one
goroutine (with associated waitGroup) it is. It's also good practice to
close all the channels to help avoid this kind of bug.

This patch also moves the waitGroup Wait into a more logical place for
visibility.
2017-11-10 14:17:54 -05:00
Jonathan Gold
c8ddbeaa5c resource: aws: ec2: Add http server 2017-11-09 13:13:42 -05:00
Jonathan Gold
3634b3450d resource: aws: ec2: Move waitgroup to resource struct 2017-11-08 16:57:41 -05:00
Jonathan Gold
c2a5e3f5d8 resources: aws: ec2: Move watch channels into struct 2017-11-08 16:16:01 -05:00
Jonathan Gold
db49fe85e4 resources: aws: ec2: Move chanStruct type out of longpollWatch 2017-11-08 16:08:25 -05:00
Jonathan Gold
567a2e9fd8 resources: aws: ec2: Reorganized consts 2017-11-08 16:02:29 -05:00
Jonathan Gold
987de00e17 resources: aws: ec2: Remove extra wait from Watch
There were two calls to WaitUntilInstanceTerminatedWithContext in a row.
There's no reason to make the call twice.
2017-11-08 16:02:24 -05:00
Jonathan Gold
baeafec74a resources: aws: ec2: Move Watch to longpollWatch 2017-11-08 16:02:12 -05:00
James Shubin
9cfa0b14d4 yamlgraph: Improve error output
This makes it easier to know what's missing.
2017-11-02 09:13:27 -04:00
James Shubin
948ded6792 github: This event is over
And it wasn't successful at all.
2017-11-01 07:07:14 -04:00
James Shubin
3c69619fd9 github: Add new label for design discussions and trackers
Open ideas related to designs can be tracked here. We've already got a
few such tickets open.
2017-11-01 07:04:32 -04:00
Jonathan Gold
e7c4bc7f47 resources: Add UserData field to AwsEc2
UserData specifies first-launch bash and cloud-init commands. See
https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/user-data.html
for documentation and examples.
2017-10-30 00:22:30 -04:00
Jonathan Gold
277ecc901b etcd: Plumbed in the new cli flags for advertise urls 2017-10-29 17:16:51 -04:00
Jonathan Gold
0f70c31a30 etcd: Add advertise urls to cli
This patch adds the option to specify URLs to advertise for clients and peers.
This will facilitate etcd communication through nat, where we want to listen
on a local IP, but expose a public IP to clients/peers.
2017-10-28 22:42:27 -04:00
James Shubin
9a97a92e31 github: Use third-party settings app to sync github settings
Let's give this a try. One downside is that giving anyone push access
gives them ability to rename repo and do other bad admin type things.
2017-10-26 05:04:41 -04:00
James Shubin
f9d452ad2c examples: Add longpoll server and client
This is an example of a race-free long-poll server and client. It uses a
redirection method to signal that the "Watch" is running.

Other race-free methods exist.
2017-10-24 04:20:19 -04:00
Jonathan Gold
9907c12eda resources: Enhancements to user and group
This patch adds autoedges between users and groups, and extends
users with additional fields for supplementary groups and a named
primary group. Also, some small fixes to log and error messages.
2017-10-23 19:18:52 -04:00
Jonathan Gold
19533a32b5 resources: Add a group resource 2017-10-21 01:28:22 -04:00
Jonathan Gold
c5a5004f9e resources: Fix user gid compare 2017-10-19 06:58:31 -04:00
Jonathan Gold
677cdea99d resources: Improve nspawn resource 2017-10-17 19:23:04 -04:00
Jonathan Gold
4d7c0ddbce resources: Add an Aws resource 2017-10-09 04:05:13 -04:00
James Shubin
81daf10157 test: Fix linter issues
These are some linter issues that were found in a new version of the
linter. Let's fix them now before that linter hits our test suite.
2017-09-26 19:38:53 -04:00
James Shubin
b3ef4e41bf test: Use stable version of gometalinter
Hopefully this prevents the various breakages seen in our lint test.
2017-09-26 19:08:43 -04:00
James Shubin
9fbf149717 etcd: Bump to newer versions 2017-09-19 18:21:15 -04:00
James Shubin
95cb94a039 vendor: Add codec package because of breakage
Recent git master 54210f4e076c57f351166f0ed60e67d3fca57a36 of
github.com/ugorji/go broke the builds. See:
https://github.com/coreos/etcd/issues/8579
2017-09-19 18:21:15 -04:00
Juan Luis de Sousa-Valadas Castaño
21f7f87716 resources: Refresh packagekit cache before install
Fixes #80
2017-09-17 22:29:15 +02:00
Jonathan Gold
831c7e2c32 resources: Add user resource 2017-09-17 01:04:36 -04:00
James Shubin
cc0d04c8b7 git: Ignore .envrc file from direnv
Some find this useful for setting a custom GOPATH per project.
2017-09-15 16:17:40 -04:00
James Shubin
46be83f8f7 legal: Re-license to GPLv3 2017-09-11 18:07:47 -04:00
James Shubin
28560e2045 resources: Fix formatting 2017-09-11 18:06:34 -04:00
James Shubin
0df4824a56 test: Increase timeouts for slow travis
Should prevent more intermittent failures.
2017-09-09 15:31:06 -04:00
James Shubin
dbcabc6517 github: Improve the PR template 2017-09-09 15:03:53 -04:00
Jonathan Gold
69f479b67e virt: Allow more than 26 disks 2017-09-08 02:15:40 +00:00
James Shubin
af75696018 github: Add a PR template to help new users
Hopefully this addresses the most common things.
2017-09-07 16:14:11 -04:00
Arthur Mello
80b8f8740f virt: Added support for ~user into expandHome
- Enabled expandHome to expand both ~/ and ~username/ paths
- Added some unit tests for expandHome
2017-09-06 14:59:08 -04:00
James Shubin
71ab325940 yaml2: Meta should keep defaults, and Res should have kind
This would previously panic since it wouldn't get a kind, and the meta
parameters would overwrite the defaults so it would block because limit
didn't have the default of +inf.

The removal of the SetKind was my fault in:

b8ff6938df

It's funny because it ends in `bad`. Guess I should have checked that!
2017-09-06 13:44:21 -04:00
James Shubin
653c76709a test: Fix another intermittent failure
Some of the tests had very precise timeouts, which weren't very
important. Here's another one that timed out early.
2017-09-04 16:39:01 -04:00
Juan Luis de Sousa-Valadas Castaño
83cc1bab38 vagrant: Fix PATH
gometalinter failed because it's not in $PATH
2017-09-04 22:08:59 +02:00
James Shubin
6c8588c019 test: Increase timeouts because travis is slow
Should hopefully prevent some intermittent failures.
2017-09-04 13:02:05 -04:00
Ismael Puerto
5b00ed2fb2 vagrant: Change box to F26
F26 provides GO 1.8
2017-09-01 22:21:39 +02:00
Juan-Luis de Sousa-Valadas Castaño
9f66962bfb docs: Change go required version to 1.8 2017-08-31 23:56:16 +02:00
James Shubin
0edba74091 etcd: Bump to version 3.2.6 and update all the grpc deps
Note: When go-grpc-prometheus was in the main $gopath (even at this
version) and everyone else was where they always were in vendor/ this
didn't build! It gave errors like:

	have SendHeader("github.com/purpleidea/mgmt/vendor/google.golang.org/grpc/metadata".MD) error
	want SendHeader("google.golang.org/grpc/metadata".MD) error

and I got frustrated. Putting it "next" to the other vendored deps seems
to have fixed this. Where are the golang docs that explain this
phenomenon?

This also requires golang 1.8+ as that is a requirement for etcd. It's
probably a reasonable thing for us too.

Note the older versions of etcd had some bugs with the concurrency
package and other things, so this is a necessary bump.
2017-08-30 14:16:02 -04:00
Dennis Kliban
1003b49dd9 resources: Add validation for Msg Priority field
This adds validation that ensures that Msg Priority field is one of the following values:
"Emerg", "Alert", "Crit", "Err", "Warning", "Notice", "Info", "Debug".
2017-08-20 12:37:39 +00:00
James Shubin
884ba54f96 resources: Include default MetaParams so Validate will pass in tests 2017-08-18 19:52:02 -04:00
Dennis Kliban
cf2325a2da vagrant: Increase amount of RAM allocated to boxes backed by libvirt 2017-08-07 13:55:21 -04:00
AdnanLFC
db6972638d pgraph: test: Added tests for DeleteEdge 2017-07-28 02:02:22 +02:00
James Shubin
74e04e81d5 travis: Update to golang 1.8 as the default
Since the release of Fedora 26 with golang 1.8.1, this is a fine
default.
2017-07-19 12:15:54 -04:00
James Shubin
7c5d7365c7 readme: Add new recording 2017-06-29 13:14:25 -04:00
James Shubin
0dadf3d78a resources: Add NewNamedResource helper
This makes the common pattern of NewResource, SetName, easier. It also
makes it less likely for you to forget to use SetName.
2017-06-17 18:09:49 -04:00
James Shubin
e341256627 resources: Add a utility to map from struct fields
For GAPI front ends that want to know what fields they can use and which
they map to, these two functions can be used.
2017-06-17 11:49:30 -04:00
James Shubin
5a3bd3ca67 hcl: Consistent formatting
Nit picks.
2017-06-16 23:01:46 -04:00
ChrisMcKenzie
8102e0a468 hcl: Added hil string interpolation to hcl frontend 2017-06-15 22:53:55 -07:00
ChrisMcKenzie
7d55179727 hcl: Removed edge object in favor of depends_on field in resource 2017-06-12 10:44:13 -07:00
ChrisMcKenzie
bc1a1d1818 hcl: Added basic hcl frontend 2017-06-09 10:31:34 -07:00
James Shubin
a8bbb22fe8 resources: Fix golint issues
Including a trick to get the golinter to allow our compact code!
2017-06-08 04:38:25 -04:00
James Shubin
6b489f71a1 remote: Add a Ready method to know when startup is finished
Previously, there was an extremely rare race where we would startup,
kick off the Run method in a goroutine, and then run Exit before Run got
very far in its execution. If Run ran some early sections of its code
_after_ we had Exited, we would trigger a panic due to the converger UID
being unregistered.

This patch blocks Exit from progressing until Run has started and
finished running. It also adds a Ready method so that you can monitor
this signal yourself if you'd like to add the necessary wait to your
code.
2017-06-08 03:55:03 -04:00
James Shubin
f1db088af4 test: Don't be noisy when running cd during testing 2017-06-08 01:05:58 -04:00
James Shubin
6fe12b3fb5 resources: Compare grouped resources properly
When comparing resources, we have to recursively compare grouped
resources as well! Now fixed.
2017-06-08 01:05:58 -04:00
James Shubin
dacbf9b68d resources: Add resource sorting and clean tests
Resource sorting is needed for comparing resource groups.
2017-06-08 01:05:58 -04:00
James Shubin
9f5057eac7 resources: Do not panic on autogrouped graph switches
Graph changes from autogrouped -> not autogrouped or vice versa cause a
panic (or I assume a leak) because we compared the auto grouped graph to
the ungrouped one, which would cause an Exit on an unstarted Vertex.
This includes a test that seems to reliably reproduces the issue.
2017-06-08 01:05:58 -04:00
James Shubin
525cd54921 pgraph: Improve testing and refactor out some test utilities 2017-06-07 07:13:12 -04:00
James Shubin
7ac94bbf5f resources: Panic if attempting to register a duplicate resource
Don't silently let this overwrite pass. It would mean a mistake.
2017-06-07 03:15:06 -04:00
James Shubin
b8ff6938df resources: Unify resource creation and kind setting
This removes the duplication of the kind string and cleans up things for
resource creation.
2017-06-07 03:07:02 -04:00
James Shubin
2f6c77fba2 misc: Update my tag script to deal with large releases 2017-06-03 03:54:49 -04:00
James Shubin
28a6430778 test: Add gometalinter to our test suite
Add a bunch of new linters to our tests! We can uncomment each sub
linter as we fix up the few remaining issues.
2017-06-03 02:04:10 -04:00
James Shubin
6e4157da35 test: Remove debugging echo from go vet test
I accidentally left it in which totally defeats the point of tests!
2017-06-03 01:34:02 -04:00
James Shubin
4f420dde05 etcd: Wait for server to start before continuing
I think there was a rare race where we would make use of the etcd server
before it had fully started up. I only ever saw this occur on travis,
and with this fix hopefully we'll never see it again.

It is worth mentioning that much of my etcd code and the lib Run()
function could use a solid cleaning.
2017-06-03 01:00:35 -04:00
James Shubin
d9601471df etcd: Small cleanup of the package
Split things into multiple files, and fix up some doc formatting.
2017-06-03 00:34:58 -04:00
James Shubin
9941a97e37 resources: pkg: Add a simple test based on internal logic
We expect the following to stay true. This has always been a bit weird
for me to either remember or expect, so I added a test for my sanity.
2017-06-03 00:15:30 -04:00
James Shubin
0a64b08669 resources: autoedges: Process in a deterministic order
The order you loop through map's isn't necessarily stable, so make sure
you sort everything before you go through it.
2017-06-02 22:29:42 -04:00
James Shubin
4d9d0d4548 resources: Improve AutoEdge API and pkg breakage
I previously broke the pkg auto edges because the package list wasn't
available by the time it was called. This fixes the pkg resource so that
it gets the necessary list of packages when needed. Since this means
that a possible failure could happen, we also update the AutoEdges API
to support errors. Errors can only be generated at AutoEdge struct
creation, once the struct has been returned (right before modification
of the graph structure) there is no possibility to return any errors.

It's important to remember that the AutoEdges stuff gets called because
the Init of each resource, so make sure it doesn't depend on anything
that happens there or that gets cached as a result of Init.

This is all much nicer now and has a test too :)
2017-06-02 22:15:28 -04:00
James Shubin
5f6c8545c6 resources: Replace stored pgraph with mgraph and clean up hacks
Now that we're using our meta wrapper graph struct instead of the
pgraph, we can re-implement our SetValue hacks in terms of struct fields
and the implementation is now cleaner.
2017-06-02 18:50:23 -04:00
James Shubin
ddc335d65a resources: Reorganize package and split into multiple files
This should hopefully make finding and changing code easier.
2017-06-02 18:08:47 -04:00
James Shubin
9cbaa892d3 gapi: Allow the GAPI implementer to specify fast and exit
This allows the implementer of the GAPI to specify three parameters for
every Next message sent on the channel. The Fast parameter tells the
agent if it should do the pause quickly or if it should finish the
sequence. A quick pause means that it will cause a pause immediately
after the currently running resources finish, where as a slow (default)
pause will allow the wave of execution to finish. This is usually
preferred in scenarios where complex graphs are used where we want each
step to complete. The Exit parameter tells the engine to exit, and the
Err parameter tells the engine that an error occurred.
2017-06-02 04:03:10 -04:00
James Shubin
9531465410 test: Make sure our examples build
Since there are occasional API changes, I'd like to at least remember to
keep the examples building, so we now have a test to remind us!
2017-06-02 03:32:53 -04:00
James Shubin
c35916fad1 resources: Rename the Data struct to ResData to avoid ambiguity
There's a similarly named gapi.Data struct which we could also rename.
2017-06-02 02:53:53 -04:00
James Shubin
bf476a058e resources: exec: Add send/recv for exec output, stdout and stderr
This adds send/recv output parameters from exec for stdout, stderr, and
output which is a combination of those two. This also includes a few
tests, and a working example too!

Gone are the `some_command > some_file` days of puppet.
2017-06-02 02:52:03 -04:00
James Shubin
d4e815a4cb resources: Clean up converger and make it easier for tests
This cleans up the resource converger code slightly and makes it easier
to write resource specific test cases.
2017-06-02 01:15:25 -04:00
James Shubin
0545c4167b pgraph: Remove NewVertex and NewEdge methods and fix examples
Since the pgraph graph can store arbitrary pointers, we don't need a
special method to create the vertices or edges as long as they implement
the String() string method. This cleans up the library and some of the
examples which I let rot previously.
2017-05-31 18:04:58 -04:00
James Shubin
6838dd02c0 resources: graph: Add partial implementation of a graph resource
This is something I've wanted to do for a while, but for the reasons
mentioned in the comments, I've been unable to complete yet. I figured
I'd at least merge what does exist so far in case someone else would
like to pick this up. It's a bit of a brain hurdle / monster, because
the tricky part is refactoring the core engine so that this fits in
nicely. Perhaps someone will have more time and/or less tunnel vision
than I to either merge something or sketch out some ideas on the path
forwards. I think it's a useful goal because if recursive resources are
possible, it could force the core engine into a more elegant design.

Happy hacking!
2017-05-31 17:27:34 -04:00
James Shubin
14c2fd1edd resources: Add proper edge compare method
Might as well do this cleanly in one place.
2017-05-31 17:27:34 -04:00
James Shubin
6e503cc79b resources: Simplify the resource Compare functions
This removes one level of indentation and simplifies the code.
2017-05-31 17:27:34 -04:00
James Shubin
bd4563b699 pgraph: Add sort function to sort a list of vertices
With tests too!
2017-05-31 17:27:34 -04:00
James Shubin
458e115490 pgraph: Add logic functions for adding subgraphs
These are helper functions to merge in existing graphs into a main graph
with or without adding an edge relationship between a vertex and the new
graph. These are particularly useful if using mgmt as a lib to break
apart units of work into functions that create sub graphs, which are
then added to the main graph when they're returned.
2017-05-31 17:27:25 -04:00
James Shubin
51369adad1 pgraph: Add a GraphCmp method
This could probably be more efficient using a known algorithm, and it
could definitely require more tests, but is good enough for now.
2017-05-31 16:45:39 -04:00
James Shubin
f65c5fb147 resources: nspawn: Fix small style issues 2017-05-31 15:36:15 -04:00
James Shubin
4150ae7307 pgraph: Replace edge struct with interface
This further cleans up the pgraph lib to be more generic.
2017-05-31 15:36:15 -04:00
James Shubin
a87288d519 pgraph, resources: Major refactoring continued
There was simply some technical debt I needed to kill off. Sorry for not
splitting this up into more patches.
2017-05-31 15:36:14 -04:00
James Shubin
3cf9639e99 pgraph, resources: Major refactor to remove pgraph to resource dep
This is the mechanical port of the remaining bits. Next to clean it up a
bit.
2017-05-29 15:43:50 -04:00
James Shubin
4490c3ed1a resources: Map to semaphores doesn't need to be a pointer
A map in golang is a reference type.
2017-05-29 15:43:50 -04:00
James Shubin
fbcb562781 pgraph: Move the timestamp storage into the resource 2017-05-29 15:43:50 -04:00
James Shubin
b1e035f96a pgraph: Move get/set state methods out to resource package 2017-05-29 15:43:50 -04:00
James Shubin
11c3a26c23 pgraph: Move the AutoEdges mechanism into the resource package
Remove the pgraph->resource dependency.
2017-05-29 15:43:50 -04:00
James Shubin
1fbe72b52d test: Run go vet across whole packages not individual files
The golang tooling is quite deficient, in that it makes it quite
difficult to get the tools to do_the_right_thing, without ample wrapping
of bash scripting. Go vet was finding issues because it didn't have the
full context available. Hopefully this package level context is
sufficient for now. It still lacks inter-package context though.
2017-05-29 15:43:50 -04:00
James Shubin
f4bb066737 test: Run go vet with -source flag in newer releases
This should hopefully eliminate some false positives.
https://github.com/golang/go/issues/20514
2017-05-29 15:43:50 -04:00
Julien Pivotto
aaac9cbeeb vagrant: Setup Packagekit in the box
Without packagekit the 'pkg' resources can not be used

Signed-off-by: Julien Pivotto <roidelapluie@inuits.eu>
2017-05-17 09:54:23 +02:00
Julien Pivotto
0e68ff6923 vagrant: Install make in the Vagrant box
Signed-off-by: Julien Pivotto <roidelapluie@inuits.eu>
2017-05-17 06:41:43 +02:00
James Shubin
1c59712cbf pgraph: Move AssociateData function out of the package
This removes another dependency on the resource package.
2017-05-15 10:19:46 -04:00
James Shubin
c2cb1c9168 pgraph: Move GraphMetas function out of package
This removes a dependency on the resources package which wasn't
necessary.
2017-05-15 10:06:31 -04:00
James Shubin
cc8e2e40dd pgraph: Update graph API to remove Get prefix and add Adjacency
Simple cleanups.
2017-05-15 09:58:10 -04:00
James Shubin
e67d97d9da pgraph: Replace CompareMatch with VertexMatchFn
This removes a reference to the resources package in pgraph.
2017-05-13 13:55:42 -04:00
James Shubin
d74c2115fd pgraph: Untangle the semaphore code from the pgraph implementation
This re-implements the semaphore code on top of the graph kv store.
2017-05-13 13:28:41 -04:00
James Shubin
70e7ee2d46 pgraph: Remove use of Flags struct in favour of Value API
One small step to completely cleaning up the pgraph package so that we
can eventually fix the code that would otherwise create a cycle!
2017-05-13 13:28:41 -04:00
James Shubin
d11854f4e8 pgraph: Clean up pgraph module to get ready for clean lib status
The graph of dependencies in golang is a DAG, and as such doesn't allow
cycles. Clean up this lib so that it eventually doesn't import our
resources module or anything else which might want to import it.

This patch makes adjacency private, and adds a generalized key store to
the graph struct.
2017-05-13 13:28:41 -04:00
James Shubin
4bb553e015 pgraph: Use the correct vertex handle to prevent a race
Small typo made that is now fixed! These need to get caught with golint!
2017-05-13 10:08:38 -04:00
James Shubin
0af9af44e5 etcd, resources, world: Add World API for shared keys
It's up to the end user to decide who is writing and/or overwriting
them.

It could also be useful to reimplement (refactor) some of the existing
World API's to be implemented in terms of these primitives.
2017-04-17 07:03:29 -04:00
James Shubin
3a0d73f740 readme: Add new links 2017-04-13 04:35:59 -04:00
James Shubin
9b9ff2622d resources: Make resource kind and baseuid fields public
This is required if we're going to have out of package resources. In
particular for third party packages, and also for if we decide to split
out each resource into a separate sub package.
2017-04-11 01:52:21 -04:00
James Shubin
a4858be967 lib, gapi: Next method of GAPI should generate first event
This puts the generation of the initial event into the Next method of
the GAPI. If it does not happen, then we will never get a graph. This is
important because this notifies the GAPI when we're actually ready to
try and generate a graph, rather than blocking on the Graph method if we
have a long compile for example.

This is also required for the etcd watch cleanup.
2017-04-10 03:20:58 -04:00
James Shubin
6fd5623b1f gapi: Move separate etcd Watch method into GAPI
This cleans up the API to not have a special case for etcd anymore. In
particular, this also adds the requirement that the GAPI must generate
an event on startup as soon as it is ready to generate a graph.
2017-04-10 03:20:58 -04:00
James Shubin
66d9c7091c lib: examples: Update to most recent API
At some point in the past the API changed. Fixed now.
2017-04-10 03:20:58 -04:00
Mildred Ki'Lya
525a1e8140 yamlgraph: Refactor parsing for dynamic resource registration
Avoid use of the reflect package, and use an extensible list of registred
resource kinds. This also has the benefit of removing the empty VirtRes and
AugeasRes struct types when compiling without libvirt and libaugeas.
2017-03-24 22:38:06 +01:00
James Shubin
64dc47d7e9 misc: Fixup documentation 2017-03-20 17:11:51 -04:00
James Shubin
f3fc7bb91e resources: svc: Add basic support for user services
These are user specific services and are available on the session bus.
This doesn't use the private user API because
https://github.com/coreos/go-systemd/pull/225 was NACKed.
2017-03-17 10:15:02 -04:00
James Shubin
028ef14cc0 misc: Replace sloppy use of %v with %s 2017-03-16 13:18:36 -04:00
James Shubin
3e001f9a1c main: Update log messages for consistency 2017-03-16 13:14:50 -04:00
Julien Pivotto
33d20ac6d8 prometheus: Add detailed metrics
Signed-off-by: Julien Pivotto <roidelapluie@inuits.eu>
2017-03-16 14:18:46 +01:00
James Shubin
660554cc45 todo: Update the TODO file to be more current
We should really try to remember to patch it with fixes when we do them
:)
2017-03-15 14:57:29 -04:00
James Shubin
a455324e8c examples: Add missing file example
I was using this for testing graph changes but forgot to commit it.
2017-03-13 08:05:49 -04:00
James Shubin
cd5e2e1148 pgraph: Add fast pausing and exiting of graphs
This causes a graph to actually stop processing part way through, even
if there are poke's that want to continue on. This is so that the user
experience of pressing ^C actually causes a shutdown without finishing
the graph execution. It might be preferred to have this be a user
defined setting at some point in the future, such as if the user presses
^C twice. As well, we might want to implement an interrupt API so that
individual resource execution can be asked to bail out early if
requested. This could happen on a third ^C press.
2017-03-13 07:54:03 -04:00
James Shubin
074da4da19 pgraph, resources: Run the resource Setup in parallel
This is a reasonable thing to do at this time.
2017-03-13 07:54:03 -04:00
James Shubin
e4e39d820c pgraph: semaphore: Refactor semaphore size function and test 2017-03-13 07:49:29 -04:00
James Shubin
e5dbb214a2 pgraph: Move the BackPoke to before the semaphores
I can't think of a reason we should grab a semaphore before backpoking.
The semaphore is intended to block around the actual work in CheckApply,
not the dependency resolution of the correct vertex.
2017-03-13 07:49:29 -04:00
James Shubin
91af528ff8 pgraph: Move the quiesce done indicator to avoid deadlock
This avoids a deadlock on resource failure when retry==0. Without this
we would never exit. This adds a test in too!
2017-03-12 13:52:35 -04:00
James Shubin
18c4e39ea3 resources: exit: Misc cleanups
Some of this code hadn't been touched much since an early mgmt. Here's a
quick cleanup of some cruft.
2017-03-12 13:21:22 -04:00
James Shubin
bda455ce78 resources: exec: Ignore signals sent to main process
When we send a ^C to the main process, our children see it too! This
puts them in their own process group so that they're not affected.
There's still the matter of properly hooking up the internal exit signal
to a proper shutdown, but that's separate.

This might mean that there should be a case for an interrupt aspect to
the resource API which would allow a second ^C by the engine, to cause a
forceful termination by the resource if that resource supported that.
2017-03-12 13:11:54 -04:00
James Shubin
a07aea1ad3 resources: exec: Clean up command error processing
Show the exit status on error and general cleanups.
2017-03-12 12:44:03 -04:00
James Shubin
18e2dbf144 resources: exec: Remove state checks that are done in the engine
These state checks are now done automatically in the engine, and so they
should be removed to make the code easier to read.
2017-03-12 12:35:03 -04:00
James Shubin
564a07e62e resources: exec: Don't invalidate state on poke
This was some legacy incorrect decision from earlier mgmt.
2017-03-12 12:35:02 -04:00
James Shubin
a358135e41 resources: exec: Remove the pollint parameter
Since we now have a poll metaparameter, we don't need the resource
specific code.
2017-03-12 10:49:26 -04:00
James Shubin
6d9be15035 pgraph: semaphore: Add lock around semaphore map
I forgot about the `concurrent map write` race, but now it's fixed. I
suppose we could probably pre-create all semaphores in the graph at once
before Start, and remove this lock, but that's an optimization for a
later day.
2017-03-11 09:06:18 -05:00
James Shubin
b740e0b78a git: Add more features to tag.sh script
This helps me make releases and probably won't help you, but why not be
transparent about things and tools!
2017-03-11 08:41:56 -05:00
James Shubin
9546949945 git: Improve tagging script 2017-03-11 08:29:53 -05:00
James Shubin
8ff048d055 test: Disable prometheus-3.sh test temporarily
It seems to be failing, and I'm not sure where the regression is, or if
there is a race. Sorry roidelapluie.
2017-03-09 11:46:21 -05:00
James Shubin
95a1c6e7fb pgraph, resources: Discard BackPokes during pause and resume
This prevents some nasty races where a BackPoke could arrive on a paused
vertex either during a resume or pause operation. Previously we might
also have poked an excessive number of resources on resume.

The solution was to discard BackPokes during pause or resume. On pause,
they can be discarded because we've asked the graph to quiesce, and any
further work can be done on resume, and on resume we ignore them because
this should only happen during the unrolling (reverse topological resume
of the graph) and at the end of this the indegree == 0 vertices will
initiate a series of pokes which should deal with any BackPoke that was
possibly discarded.

One other aspect of this which is important: if an indegree == 0 vertex
is poked (Process runs) but it's already in the correct state, it should
still transmit the Poke through itself so that subsequent vertices know
to run. Currently this is done correctly in Process().

I'm a bit ashamed that this wasn't done properly in the engine earlier,
but I suppose that's what comes out of running fancier graphs and really
thinking in detail about what's truly correct. Hopefully I got it right
this time!
2017-03-09 06:35:15 -05:00
James Shubin
0b1a4a0f30 pgraph, resources: Quiesce when pausing or exiting the resource
This prevents a nasty race that can happen in a graph with more than one
resource. If a resource has someone that it can BackPoke, and then
suppose an event comes in. It runs the obj.Event() method (from inside
its Watch loop) and then *before* the resulting Process method can run
it receives a pause event and pauses. Then the parent resource pauses as
well. Finally (it's a race) the Process gets around to running, and
decides it needs to BackPoke. At this point since the parent resource is
paused, it receives the BackPoke at a time when it can't handle
receiving one, and it panics!

As a result, we now track the number of running Process possibilities
via a WaitGroup which gets incremented from the obj.Event() and we don't
finish our pause or exit operations until it has quiesced and our
WaitGroup lets us know via Wait(). Lastly in order to prevent repeated
replays, we detect when we're quiescing and suspend replaying until post
pause. We don't need to save the replay (playback variable) explicitly
because its state remains during pause, and on exit it would get
re-checked anyways.
2017-03-09 02:50:55 -05:00
James Shubin
22b48e296a resources, yamlgraph: Drop the kind capitalization
This stopped making sense now that we have a resource with two primary
capitals. It was just a silly formatting hack anyways. Welcome kv!
2017-03-09 02:50:55 -05:00
James Shubin
c696ebf53c resources: svc: Add failed state
Services can be in a failed state too.
2017-03-08 19:23:33 -05:00
James Shubin
a0686b7d2b pgraph: graphviz: Update Graphviz lib to quote names properly
This also moves the library to after the graph starts so that the kind
fields will be visible.
2017-03-08 19:23:33 -05:00
James Shubin
8d94be8924 resources: kv: Add new KV resource which sets key value pairs
This is a new resource for setting key value pairs in our global world
database. Currently only etcd is supported. Some of the implications and
possibilities of this resource will become more obvious with future
commits!

You can bother/test this resource with these commands:

ETCDCTL_API=3 etcdctl get "/_mgmt/strings/" --prefix=true
ETCDCTL_API=3 etcdctl put "/_mgmt/strings/KEY/HOSTNAME" 42

Replace the KEY and HOSTNAME variables with the actual values you'd like
to use. The 42 is the value that is set.
2017-03-08 19:23:33 -05:00
James Shubin
e97ac5033f resources: Split util functions into separate file
This also adds errwrap to their implementation.
2017-03-08 19:23:33 -05:00
James Shubin
44771a0049 gapi: Move the World interface into resources
This was necessary to fix some "import cycle" errors I was having when
adding the World api to the resource Data struct.

I think this is a good hint that I need to start splitting up existing
packages into sub files, and cleaning up and inter-package problems too.
2017-03-08 19:23:33 -05:00
James Shubin
32aae8f57a lib, pgraph, resources: Refactor data association API
This should make things cleaner and help avoid as much churn every time
we change a property.
2017-03-07 22:51:11 -05:00
James Shubin
8207e23cd9 lib: Refactor instantiation of world API 2017-03-07 22:51:11 -05:00
James Shubin
a469029698 etcd: Switch to buffered channel to remove duplicates
Since we don't return the actual values and instead only tell about
events (which leaves the `Get` of the value as a second operation) then
we don't have to use a channel with backpressure since all the events
are identical.
2017-03-07 22:51:11 -05:00
James Shubin
203d866643 gapi, etcd: Define and implement a string sharing API for the World
This adds a new set of methods to the World API (for sharing data
throughout the cluster) and adds an etcd backed implementation.
2017-03-07 22:51:11 -05:00
Michael Borden
1488e5ec4d travis: Add go_import_path
This should fix an issue with turning on travis for any mgmt fork.
2017-03-07 14:17:53 -08:00
Michael Borden
af66138a17 misc: Add pacman support 2017-03-03 21:04:29 -08:00
James Shubin
5f060d60a7 test: Avoid matching three X's
This helps my "WIP" detector script avoid false positives. It is a
simple script which helps me find release critical problems.
2017-03-01 22:37:08 -05:00
James Shubin
73ccbb69ea travis: Ensure recent golang versions in test
This ensures travis picks the latest versions. Apparently writing 1.x is
also a valid choice.
2017-03-01 21:45:02 -05:00
James Shubin
be60440b20 readme: Add new blog post about metaparameters
This also documents the recent semaphore work.
2017-03-01 16:18:14 -05:00
James Shubin
837efb78e6 spelling: Fix typos as found by goreportcard 2017-02-28 23:48:34 -05:00
James Shubin
4a62a290d8 pgraph: Clean up tests
This splits the tests into multiple files.
2017-02-28 16:47:16 -05:00
James Shubin
018399cb1f semaphore, pgraph: Add semaphore grouping and tests
If two resources are grouped, then the result should contain the
semaphores of both resources. This is because the user is expecting
(independently) resource A and resource B to have a limiting choke
point. If when combined those choke points aren't preserved, then we
have broken an important promise to the user.
2017-02-28 16:40:53 -05:00
James Shubin
646a576358 remote: Replace builtin semaphore type with common util lib
Refactor code to use the new fancy lib!
2017-02-27 02:57:06 -05:00
James Shubin
d8e19cd79a semaphore: Create a semaphore metaparam
This adds a P/V style semaphore mechanism to the resource graph. This
enables the user to specify a number of "id:count" tags associated with
each resource which will reduce the parallelism of the CheckApply
operation to that maximum count.

This is particularly interesting because (assuming I'm not mistaken) the
implementation is dead-lock free assuming that no individual resource
permanently ever blocks during execution! I don't have a formal proof of
this, but I was able to convince myself on paper that it was the case.

An actual proof that N P/V counting semaphores in a DAG won't ever
dead-lock would be particularly welcome! Hint: the trick is to acquire
them in alphabetical order while respecting the DAG flow. Disclaimer,
this assumes that the lock count is always > 0 of course.
2017-02-27 02:57:06 -05:00
James Shubin
757cb0cf23 test: Small fixups to t4 and a rename 2017-02-26 20:54:22 -05:00
Julien Pivotto
7d92ab335a prometheus: Add mgmt_pgraph_start_time_seconds metric
Signed-off-by: Julien Pivotto <roidelapluie@inuits.eu>
2017-02-26 15:28:43 +01:00
James Shubin
46c6d6f656 compiling: Add -i flag to go build to speed up builds
So builds take about 30s on my shitty machine which is pretty long.
Turns out golang caches things in $GOPATH/pkg/ but is not clever enough
to erase things from there that are out of date. As a result, it was
rebuilding everything (including the unchanged dependencies) every
build!

By completely wiping out $GOPATH/pkg/ and then running `go build -i`,
this now takes builds down to about 8 seconds. (After one full build is
finished.)

This is basically the same as running `go install`, but without copying
junk to $GOPATH/bin.

Hopefully the tooling will be smart enough to know when to throw out
stuff in $GOPATH/pkg automatically and avoid this problem entirely.

Is it wrong to send Google a bill for all the extra cpu cycles I've
used? ;)
2017-02-26 04:06:36 -05:00
Julien Pivotto
46260749c1 prometheus: Move the prometheus nil check inside the prometheus function
That pattern will be reused in future metrics.

Signed-off-by: Julien Pivotto <roidelapluie@inuits.eu>
2017-02-26 09:33:34 +01:00
Julien Pivotto
50664fe115 compilation: Make build the default target
It is really strange that whe I run make, it does not build mgmt. This
commit makes build the default target, without moving the target,
therefore we keep as much as we can the order of the file.

This also removes the confusion for designers that would run "make"
instead of "make art", whose work would be disrupted when we add a --
let's say -- make alpharelese command.

Signed-off-by: Julien Pivotto <roidelapluie@inuits.eu>
2017-02-26 09:04:20 +01:00
James Shubin
c480bd94db resources: virt: Remove unnecessary early exit from CheckApply
I don't think this early exit is necessary any more, since the main
CheckApply function really just spawns out to the different sub workers
which all individually check the apply variable.

If I'm wrong, we can revert this. It was @roidelapluie that noticed the
check here to begin with.
2017-02-25 21:29:08 -05:00
James Shubin
79923a939b resources: virt: Catch bad calls to CheckApply
If the engine cheats, we'll know!
2017-02-25 21:28:35 -05:00
James Shubin
327b22113a resources: virt: Don't block exit in callbacks
This prevents us blocking an exit if we close when a callback was about
to run. This is because the callbacks are called from the
EventRunDefaultImpl method, which waits for their return to exit and
release the WaitGroup.

I think we should probably get rid of the obj.wg since the engine is
supposed to guarantee that Close doesn't happen before Watch finishes.
2017-02-25 21:04:36 -05:00
James Shubin
12160ab539 pgraph: Wait for Process routine to exit
Wait for the innerWorker's Process routine to exit, before we exit too.
2017-02-25 21:01:02 -05:00
James Shubin
2462ea0892 pgraph, resources: Wait for innerWorker to exit cleanly
Don't run the Close() method until the innerWorker has exited cleanly.
This is a guarantee which we make to the resources.
2017-02-25 21:00:38 -05:00
James Shubin
8be09eadd4 test: Fix up probable timeout failures due to slow ci
We had *occasional* failures most likely due to slow ci and compounded
by low entropy. We also weren't pointing at the right test!
2017-02-22 23:08:09 -05:00
James Shubin
98bc96c911 golint: Fixup issues found in the report
This also increases the max allowed to 5% -- I'm happy to make this
lower if someone asks.
2017-02-22 22:18:55 -05:00
James Shubin
b0fce6a80d travis: Start building golang 1.8 as well
Require success from 1.7 since we'll probably move to it as the default
within six months.
2017-02-22 22:18:55 -05:00
James Shubin
53b8a21d1e resources: virt: Cleanup cleanly on Close
Don't block accidentally on error!
2017-02-22 22:18:55 -05:00
James Shubin
1346492d72 yamlgraph: Close the recwatcher properly after use
Don't leave it running unnecessarily! This might have contributed to a
block, but it was hard to isolate if this was the cause or if this was
one of many causes.
2017-02-22 18:37:43 -05:00
James Shubin
e5bb8d7992 recwatch: Close the ConfigWatch properly
This improves the shutdown process so that there is no change of
blocking if the sender runs close without having emptied the channel.
2017-02-22 17:45:16 -05:00
James Shubin
49594b8435 pgraph, resources: Clean up the event system around the resources
This cleans up some of the resource events and also reorganizes the
struct for simplicity. This should hopefully kill off at least one race
which would cause unnecessary blocking!

Yes this patch is a bit yucky, but so was the bug I was fighting with!
2017-02-22 17:45:16 -05:00
James Shubin
3bd37a7906 test: Don't block on graph transitions
Improvements in the engine have uncovered some annoying race conditions
which would cause the engine to block between transitions. This is a
test which catches the most obvious file based ones.

This requires inotify to work in the test environment.
2017-02-22 17:45:16 -05:00
James Shubin
e070a85ae0 lib: Misc cleanups and new log message 2017-02-22 17:45:16 -05:00
James Shubin
c189278e24 recwatch: Unblock from sending on exit
If we receive an exit signal while we are waiting to send a message, we
should still be able to shut down.
2017-02-22 16:41:47 -05:00
James Shubin
2a8606bd98 recwatch: Close cleanly after Watcher finishes
This cleans up the recwatch code to avoid some possible races. While you
were previously able to call Close more than once, this was really
something that would mask bugs, rather than a useful feature.
2017-02-22 16:41:26 -05:00
James Shubin
18ea05c837 pgraph, resources: Add proper start/stop signals
We need to perform some operations in lock step between graph
transitions. This should help with that!
2017-02-21 18:48:27 -05:00
James Shubin
86c3072515 resources: The user should not call Init
Even in tests...
2017-02-21 18:42:07 -05:00
James Shubin
fccf508dde resources, pgraph: Refactor Worker and simplify API
I'm still working on reducing the size of the monster patches that I
land, but I'm exercising the priviledge as the initial author. In any
case, this refactors worker into two, and cleans up the passing around
of the processChan. This puts common code into Init and Close.
2017-02-21 18:42:07 -05:00
James Shubin
2da21f90f4 pgraph, resources: Improve Init/Close and Worker status
This should do some rough cleanups around the Init/Close of resources,
and tracking of Worker function status.
2017-02-21 18:42:07 -05:00
James Shubin
bec7f1726f resources: virt: Allow hotplugging
This allows hot (un)plugging of CPU's! It also includes some general
cleanups which were necessary to support this as well as some other
features to the virt resource. Hotunplug requires Fedora 25.

It also comes with a mini shell script to help demo this capability.

Many thanks to pkrempa for his help with the libvirt API!
2017-02-21 18:42:07 -05:00
James Shubin
74dfb9d88d test: Make test status more clear 2017-02-21 18:40:31 -05:00
James Shubin
02dddfc227 test: Fix yamlfmt test
Last chance before we kill this entirely.
2017-02-21 16:16:41 -05:00
Julien Pivotto
545016b38f project: Add $me to AUTHORS
Signed-off-by: Julien Pivotto <roidelapluie@inuits.eu>
2017-02-21 07:07:41 +01:00
Felix Frank
0ccceaf226 project: Reluctantly add myself as an author
For the record, I currently don't feel that my level of contributions
warrant a spot here, but I'm entering myself upon express invitation
by James =) ...also in anticipation of much more substantial work.
2017-02-20 22:53:30 -05:00
James Shubin
a601115650 test: Fix false negative on go vet
This was my fault, now it is fixed :) It passed locally due to a bug.
2017-02-20 18:12:30 -05:00
Kaushal M
ae6267c906 build: Use GOTAGS for static builds as well 2017-02-20 17:16:03 -05:00
James Shubin
ac142694f5 test: Improve go vet so that it is less noisy 2017-02-20 17:08:48 -05:00
James Shubin
69b0913315 test: Fix tests by hooking up go test properly
The internal golang tests broke when we turned everything into packages.
This resurrects them with the hopes that we'll add more!
2017-02-20 16:40:40 -05:00
Julien Pivotto
421bacd7dc travis: Run apt update before installing dependencies
Signed-off-by: Julien Pivotto <roidelapluie@inuits.eu>
2017-02-20 22:23:06 +01:00
James Shubin
573a76eedb build: Require golang 1.6 or greater 2017-02-20 15:17:49 -05:00
James Shubin
b7948c7f40 resources: Specify defaults for the MetaParams
When creating new resources, we didn't specify the defaults, which for
the limit metaparam caused invalid resources by default. It would be
nice to change the limit param to have the 1/X (reciprocal) as the
default, although the problem with that is that (1) it is illogical, and
(2) it's not clear if the precision for the common cases is enough.

If someone wants to investigate this further, please do! Zero value
structs are definitely more useful! In any case, we can now specify the
default. It's not entirely obvious to me if this is the best way to do
it, or if there is a superior method.
2017-02-16 21:08:46 -05:00
James Shubin
2647d09b8f resources: file: Don't modify resource in Init
This didn't break anything previously, but technically wasn't correct.
Pure functions are superior in this case!
2017-02-16 21:06:58 -05:00
James Shubin
57e919d7e5 resources: Remove "NewRes" constructors
Remove the New constructors since calling Init should be done by the
engine, and not by the user even when using mgmt as a lib. This is also
the case in tests! It used to be the case that a user might want to call
Init manually, but that is no longer the case!
2017-02-16 21:06:12 -05:00
James Shubin
f456aa1407 resources: file: Small fixups and force additions 2017-02-16 20:46:51 -05:00
Mildred Ki'Lya
d0d62892c8 resources: file: Allow creation of empty directories 2017-02-16 20:41:59 -05:00
James Shubin
a981cfa053 legal: Oh yeah, it is 2017 2017-02-16 01:34:32 -05:00
James Shubin
55290dd1e3 docs: Add missing newline and remove extra one 2017-02-16 01:13:42 -05:00
Julien Pivotto
9c4e255994 documentation: Implement sphinx documentation
Licence removed due to to a read the docs (or sphinx/recommonmark) bug:
everything after the comment is not rendered.

Signed-off-by: Julien Pivotto <roidelapluie@inuits.eu>
2017-02-16 07:08:44 +01:00
Julien Pivotto
f9c7d5f7bc resources: augeas: comments: Improvements
- Remove obvious statements
- Fix typo

Signed-off-by: Julien Pivotto <roidelapluie@inuits.eu>
2017-02-15 22:42:42 +01:00
Julien Pivotto
49baea5f6a documentation: Embed command-line syntax in a code block
Signed-off-by: Julien Pivotto <roidelapluie@inuits.eu>
2017-02-15 15:30:32 -05:00
James Shubin
6209cf3933 resources: file: Use the computed path in our resource
We compute the actual path in Init() and forget to use it everywhere.
2017-02-15 15:25:03 -05:00
Mildred Ki'Lya
d170a523c3 misc: Split OPTS variable of systemd unit
Systemd split variables when specified like $VAR and not when specified
like ${VAR}. Since OPTS must contain multiple options, we must ensure
systemd will split it.

See also systemd.service(5)
2017-02-15 19:31:44 +01:00
Julien Pivotto
be5040e7a8 prometheus: Remove mgmt_process_start_time_seconds metric
That metric is useless as by default the prometheus golang client
provides the `process_start_time_seconds` metric.

This reverts commit 25e2af7c89.
2017-02-14 22:56:12 +01:00
James Shubin
ecbaa5bfc1 github: Show a friendly message in issues template 2017-02-14 16:45:35 -05:00
Julien Pivotto
25e2af7c89 prometheus: Add mgmt_process_start_time_seconds metric 2017-02-14 22:14:59 +01:00
Julien Pivotto
605688426d resources: file: Do not error on os.Stat in noop mode
Fix #142.

Signed-off-by: Julien Pivotto <roidelapluie@inuits.eu>
2017-02-14 20:54:32 +01:00
James Shubin
0e069f1e75 docs: Update the quick start guide with git clone and vagrant
There was a bit of a catch-22, since go get wouldn't succeed without
some system build libs installed first. We also added a Vagrant
environment.
2017-02-14 12:13:43 -05:00
James Shubin
e9adbf18d3 test: prometheus: Fix up test 2017-02-14 12:10:54 -05:00
Sean Jones
610202097a test: Enable macOS shell testing
* Check and install libvirt with Homebrew

  macOS does not have apt, dnf or yum. Add checking for homebrew for
  installing libvirt.

* Use platform timeout for tests
    * Add timeout detection to test/util.sh
    * Use $timeout for shell test requiring timeout
2017-02-14 11:59:44 -05:00
Mildred Ki'Lya
8c2c552164 resources: file: Implement file attributes
Add owner which must be username or uid of the file owner, group which is
the group name or gid of the file, and mode which is the octal unix file
permissions.

Add separate implementation for Go 1.6 and lower.
2017-02-14 11:55:00 -05:00
Julien Pivotto
b9976cf693 documentation: prometheus: Improvements
Signed-off-by: Julien Pivotto <roidelapluie@inuits.eu>
2017-02-13 21:59:05 +01:00
Julien Pivotto
3261c405bd resources: augeas: Make augeas support optional
Signed-off-by: Julien Pivotto <roidelapluie@inuits.eu>
2017-02-13 07:14:14 +01:00
James Shubin
35d3328e3e etcd: Remove stuttering in package
This is a good first step to cleaning up the package.
2017-02-12 22:51:46 -05:00
James Shubin
e96041d76f resources: augeas: Turn augeas namespace into a constant 2017-02-12 19:46:16 -05:00
Felix Frank
c2034bc0c0 vagrant: Add Vagrantfile and some basic configuration 2017-02-12 19:18:33 -05:00
Julien Pivotto
e8855f7621 prometheus: Implement mgmt_checkapply_total metric
Signed-off-by: Julien Pivotto <roidelapluie@inuits.eu>
2017-02-12 23:45:47 +01:00
Julien Pivotto
bdb8368e89 resources: augeas: New resource
Signed-off-by: Julien Pivotto <roidelapluie@inuits.eu>
2017-02-12 23:02:12 +01:00
Julien Pivotto
f160db2032 compilation: virt: Make libvirt support optional
refs #114

Signed-off-by: Julien Pivotto <roidelapluie@inuits.eu>
2017-02-12 22:09:46 +01:00
Julien Pivotto
de9a32a273 recwatch: Remove watcher on file move
Fix #120

Signed-off-by: Julien Pivotto <roidelapluie@inuits.eu>
2017-02-11 13:45:32 +01:00
Steve Milner
6ba7422c3b spec: Fix rpmlint warnings
- Switched spaces to tabs
- Added quiet switch to setup
2017-02-10 11:55:58 -05:00
Julien Pivotto
5cbb0ceb80 test: commit: Improve commit message testing
Signed-off-by: Julien Pivotto <roidelapluie@inuits.eu>
2017-02-10 11:12:26 +01:00
James Shubin
5b29358b37 test: Small nitpicks with messages 2017-02-09 11:16:18 -05:00
Julien Pivotto
90147f3dfb travis: more strict commit messages tests
Signed-off-by: Julien Pivotto <roidelapluie@inuits.eu>
2017-02-09 16:33:06 +01:00
Julien Pivotto
72873abe05 test: file: test the behaviour of inotify on parent dir moves
This is a test for #124. It is disabled until #124 is fixed, so it can
already me merged.

Signed-off-by: Julien Pivotto <roidelapluie@inuits.eu>
2017-02-09 16:01:09 +01:00
Julien Pivotto
de1810ba68 travis: add a test regarding commit messages
Signed-off-by: Julien Pivotto <roidelapluie@inuits.eu>
2017-02-09 16:01:09 +01:00
Julien Pivotto
7b7c765d78 prometheus: Add a new test, with --prometheus-listen
Also: rename t9 to prometheus-1

Signed-off-by: Julien Pivotto <roidelapluie@inuits.eu>
2017-02-09 00:05:19 +01:00
Felix Frank
806d4660cf tests: simplify shell code, skip YAML test if Ruby is too old 2017-02-08 09:29:00 -05:00
Julien Pivotto
5ae5d364bb Prometheus: fix URL to the "Default port allocations"
Signed-off-by: Julien Pivotto <roidelapluie@inuits.eu>
2017-02-08 13:41:09 +01:00
Julien Pivotto
1af67e72d4 prometheus: Implement basic Prometheus support
Signed-off-by: Julien Pivotto <roidelapluie@inuits.eu>
2017-02-08 12:13:33 +01:00
James Shubin
ed268ad683 pgraph, yamlgraph: Allow specifying notify value from YAML 2017-02-06 16:29:02 -05:00
James Shubin
5bdd2ca02f examples: Update the examples 2017-02-06 16:28:16 -05:00
Daniele Sluijters
eb59861d4d TODO: Point to current file improvement issue 2017-02-06 13:31:26 +01:00
Tom Ritserveldt
427e46a2aa point to the correct puppet-guide path in docs. 2017-02-06 10:35:39 +01:00
James Shubin
68a8649292 resources: Parse YAML infinity specifications correctly
This makes it easier to specify an infinite rate.
2017-02-05 21:01:52 -05:00
James Shubin
2aff8709a5 gapi: Unblock from a waiting send on GAPI close
There was a race condition that would sometimes occur in that if we
stopped reading from the gapiChan (on shutdown) but then a new message
was available before we managed to close the GAPI, then we would wait
forever to finish the close because the channel never sent, and the
WaitGroup wouldn't let us exit.

This fixes this horrible, horrible race.
2017-02-05 21:01:52 -05:00
James Shubin
62c3add888 lib: Make initial one-shot start pattern more elegant
This should do the same thing, in a more elegant and efficient way.
2017-02-05 19:00:14 -05:00
James Shubin
3ac878db62 recwatch: Do not send to channel if closed.
If we close the recwatcher right after it's opened, we might have
previously tried to send on the now closed channel.
2017-02-05 18:57:08 -05:00
James Shubin
c247cd8fea resources: Don't double close on Running restart
If we use the "retry" metaparam on a Watch, we want to avoid a double
close due to the second Running() signal. This avoids this with a simple
flag.
2017-02-05 18:47:40 -05:00
James Shubin
b6772b7280 examples: etcd: Simplify the etcd examples 2017-02-03 19:52:29 -05:00
James Shubin
807a3df9d1 travis: Limit notifications now that travis cron is enabled
This should hopefully cause less noise!
2017-02-01 08:45:57 -05:00
James Shubin
491d60e267 examples: Have libmgmt work around the lack of rate defaults
The rate limit lib needs a proper default. Until that's the case,
specify this so it doesn't block the resources.
2017-01-27 18:41:09 -05:00
James Shubin
4811eafd67 etcd: Update example to ignore pgp
This makes demos go faster since pgp is wip anyways.
2017-01-27 18:24:47 -05:00
James Shubin
8dedbb9620 examples: Update example to be Validate safe
This is also safer too!
2017-01-27 18:24:05 -05:00
James Shubin
dd8454161f pgraph: Set the false starter value too
We might have left re-used nodes as true, even if they no longer were
anymore due to graph changes, which would have caused additional pokes.
2017-01-27 17:43:24 -05:00
James Shubin
9421f2cddd resources: Rename GetUIDs to UIDs
This is more in line with the style guide for golang.
2017-01-25 14:51:23 -05:00
James Shubin
d8c4f78ec1 virt: Allow the use of ~ to expand to home directory
This makes examples slightly nicer to commit, since you don't have to
have a hardcoded ~/james/ in their source value. It's also probably a
useful feature for the resource.
2017-01-25 13:06:28 -05:00
James Shubin
54296da647 converger: Remove converger boilerplate from the resources
This simplifies the resource code by now removing all the converger
related material. Happy resource writing!
2017-01-25 11:30:47 -05:00
James Shubin
357102fdb5 converger: Block converging in the engine
This more appropriately blocks converging in the engine, since we are
now 1-1 decoupled from the Watch resource. This simplifies resource
writing, and should be more accurate around small converged timeouts.

We don't block in the Worker routine when we are polling, because we
expect to get constant poll events, and we can instead be more careful
about these by looking at CheckApply results.

If we can do this for all resources in the future, it would be
excellent!
2017-01-25 11:06:02 -05:00
James Shubin
7e15a9e181 pgraph: Add debug messages
These two messages in particular make graph analysis easier.
2017-01-25 09:52:34 -05:00
James Shubin
12e0b2d6f7 pgraph: Parallelize the BackPoke facility
I can't guarantee this has a significant effect, but it's likely to add
some efficiency when sending multiple BackPoke's at the same time, so
that erroneous ones can be cancelled out easier.
2017-01-25 09:13:59 -05:00
James Shubin
11b40bf32f resources: Update state checks
The mgmt graph depends on state tracking to eliminate redundant pokes.
With the Watch loop now able to produce events quickly, it should no
longer play a part in determining the vertex state. This simplifies the
resource API as well!
2017-01-25 09:13:59 -05:00
James Shubin
8d2b53373f pgraph: Remove unnecessary indentation in Process code path
We can indent the simple BackPoke code path instead!
2017-01-25 09:13:59 -05:00
James Shubin
9ecc49e592 resources: Force a sane default for zero value limiting
The default UnmarshalYAML on *BaseRes doesn't work properly at the
moment, so hack in a default so that we don't need to specify one if the
MetaParams struct isn't specified. The problem is that if there isn't a
meta value added, its UnmarshalYAML doesn't get a chance to run.
2017-01-25 09:13:59 -05:00
James Shubin
4f34f7083b resources: rate limiting: Implement resource rate limiting
This adds rate limiting with the limit and burst meta parameters. The
limits apply to how often the Process check is called. As a result, it
might get called more often than there are Watch events due to possible
Poke/BackPoke events.

This system might need to get rethought in the future depending on its
usefulness.
2017-01-25 09:13:59 -05:00
James Shubin
2a6df875ec resources: Improve composition of Validate API in resources
This now appropriately calls the Base method.
2017-01-22 06:00:37 -05:00
James Shubin
51c83116a2 resources: Overhaul legacy code around the resource API
This patch makes a number of changes in the engine surrounding the
resource API. In particular:

* Cleanup of send/read event.
* Cleanup of DoSend (now Event) in the Watch method.
* Events are now more consistently pointers.
* Exiting within Watch is now done in a single place.
* Multiple incoming events will be combined into a single action.
* Events in flight during an action are played back after CheckApply.
* Addition of Close method to API

This gets things ready for rate limiting and semaphore metaparams!
2017-01-22 05:59:15 -05:00
James Shubin
74435aac76 docs: Small fixes to the README 2017-01-22 05:47:42 -05:00
goberghen
5dfdb5b5f9 docs: Fix headers formating 2017-01-22 16:05:27 +06:00
James Shubin
ac892a3f3d docs: Add missing newline 2017-01-16 11:15:59 -05:00
James Shubin
1a2e99f559 docs: Add new blog post link 2017-01-16 11:14:19 -05:00
James Shubin
e97bba0524 docs: Add an introductory paragraph to the quick start guide 2017-01-16 11:12:07 -05:00
James Shubin
0538f0c524 docs: Improve README 2017-01-16 11:01:27 -05:00
James Shubin
fc3e35868d docs: Rename puppet guide and small cleanups 2017-01-16 10:27:17 -05:00
James Shubin
f1e0cfea1c docs: Split out a separate quick start guide 2017-01-16 10:27:17 -05:00
James Shubin
56efef69ba resources: Officially add Validate method
This officially adds the Validate method to the resource API, and also
cleans up the ordering in existing resources.
2017-01-09 05:10:26 -05:00
James Shubin
668ec8a248 docs: Add missing headings to table of contents 2017-01-09 04:35:48 -05:00
James Shubin
60912bd01c resources: Add a Default method to the resource API
This provides sensible defaults for when they're not the zero value.
2017-01-09 04:35:48 -05:00
Daniel P. Berrange
0b416e44f8 resources: virt: convert to use github.com/libvirt/libvirt-go
Convert the code to use the new libvirt Go bindings.

Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
2017-01-05 04:55:51 -05:00
James Shubin
ecc4aa09d3 resources: timer: Polish the timer resource
We should probably use this more correct data type which now matches
what is used by the Poll metaparam.
2016-12-24 00:51:39 -05:00
James Shubin
b921aabbed resources: Add poll metaparameter
This allows a resource to use polling instead of the event based
mechanism. This isn't recommended, but it could be useful, and it was
certainly fun to code!
2016-12-24 00:51:39 -05:00
James Shubin
6ad8ac0b6b converger: Wait till the timer exits
Wait till the timer loop exits, otherwise we might race and panic if
ConvergedTimer gets called after we've asked for an exit, but before it
does.
2016-12-24 00:51:39 -05:00
James Shubin
44e7e0e970 resources: virt: Add missing converger call
This was previously left out accidentally I suppose.
2016-12-23 23:33:13 -05:00
James Shubin
45820b4ce3 resources: Rename Converger to ConvergerUID
This is more correct since we need the Converger method to return the
actual converger!
2016-12-23 23:32:05 -05:00
Lars Kulseng
3a098377cb readme: Changed wording of quick start section 2016-12-23 23:29:27 +01:00
James Shubin
35875485ee todo: Make TODO file more current 2016-12-21 03:42:44 -05:00
James Shubin
19760be0bc golint: Fix some golint issues 2016-12-21 03:10:25 -05:00
James Shubin
b3ea33f88d test: Allow devel versions to run gofmt
Let tip builds pass in travis too!
2016-12-21 02:48:50 -05:00
James Shubin
5b3425a689 pgraph: Remember to unpause the vertices!
Forgot this part earlier, sorry! Should work correctly now :)
2016-12-21 02:39:54 -05:00
James Shubin
a3d157bde6 pgraph: Mutex must be a pointer
This should be done the same way as the WaitGroup so that we don't
panic!
2016-12-20 05:49:17 -05:00
James Shubin
2c8c9264a4 pgraph: Simplify graph exit waiting
I think the vertex resource exiting can be done in a single stage
instead of the previous two stage exit.
2016-12-20 05:49:17 -05:00
James Shubin
0009d9b20e pgraph, resources: Integrate properly with the startup logic
This signals which resources have to run their initial pokes, and
removes the racy retry timer. We actually get a proper signal when
things are running too!
2016-12-20 05:49:17 -05:00
James Shubin
dd8d17232f pgraph: Build the sync group into the graph structure
This hides the sync/wait logic inside the graph itself.
2016-12-20 05:49:17 -05:00
James Shubin
6312b9225f gapi: Rename SwitchStream to Next
This is more concise and I think more logical. Complains welcome!
2016-12-20 05:49:17 -05:00
James Shubin
68cc09fef2 resources: file: Fix small typo in the compare function 2016-12-20 05:49:16 -05:00
James Shubin
0651c9de65 docs: Add resource guide
Sorry I never published this earlier. Thanks to everyone who has managed
to write a native resource without this.
2016-12-20 05:49:16 -05:00
James Shubin
38261ec809 resources: msg: Remove legacy comment
This doesn't apply anymore. Remove it!
2016-12-20 05:47:40 -05:00
James Shubin
067932aebf resources: Remove SetWatching/IsWatching code from Watch
This removes some boilerplate from the Watch methods which can be baked
into the engine instead.

This code should be checked for races and locks to make sure we only
start resources when it makes sense to.
2016-12-20 05:47:40 -05:00
James Shubin
af47511d58 resources: Don't dirty resource when poked with activity
When we receive a poke with the activity flag set it probably means we
are receiving a refresh notification. This doesn't necessarily mean that
the resource state should be dirty as a result, in particular if the
resource doesn't support refreshing. As a result, don't automatically
mark it as dirty. (The engine knows not to skip the CheckApply when the
refresh flag is set!)
2016-12-20 05:47:40 -05:00
James Shubin
36b916f27f resources: Simplify resource Converger and Startup code
This takes the Converged initialization and Startup patterns that are
common in all resources, and bakes it into the core engine. This way
resource writing is much more concise and there is less boilerplate!
2016-12-20 05:47:40 -05:00
James Shubin
e519811893 docs: Create a dedicated documentation folder 2016-12-09 17:32:50 -05:00
James Shubin
4803be1987 misc: Rename mgmtmain to lib and remove global package
This refactor should make it cleaner to use mgmt.
2016-12-08 23:31:45 -05:00
James Shubin
1f415db44f readme: Add new blog post about send/recv 2016-12-07 14:22:01 -05:00
James Shubin
0e316b1d55 gapi: Add world interface and refactor existing code to use it
This is the initial base of what will hopefully become a powerful API
that machines will use to communicate. It will be the basis of the
stateful data store that can be used for exported resources, fact
exchange, state machine flags, locks, and much more.
2016-12-07 02:39:14 -05:00
James Shubin
eb545e75fb resources: Re-order send/recv display messages.
The updated order is more logical, and follows the time sequence.
2016-12-06 14:42:06 -05:00
James Shubin
6edb5c30d5 resources: Actually verify which send/recv elements changed
When updating the code, I forgot to actually verify if there were
changes or not. This caused erroneous changed messages when none were
actually sent.
2016-12-06 14:22:34 -05:00
James Shubin
597ed6eaa0 resources: Polish the password PoC and build out send/recv
This polishes the password resource so that it can actually avoid
writing the password to disk, and so that the work actually happens in
CheckApply where it can properly interact with the graph. This resource
now re-generates the password when it receives a notification.

The send/recv plumbing has been extended so that receivers can detect
when they're receiving new values. This is particularly important if
they might otherwise not expect those values to change and cache them
for efficiency purposes.
2016-12-06 02:29:47 -05:00
Nicolas Nadeau
2b47d7494e pgp: Base pgp code 2016-12-05 02:10:15 -05:00
James Shubin
213a88f62f misc: Improve gofmt test case
Add new golang versions, and fail if one is not found.
2016-12-04 21:11:36 -05:00
James Shubin
07fd2e88a2 resources: Fix poke/refresh race
Clearly the use of errgroup is flawed.
1) You can't pass in variables, so this is likely to race.
2) You can't get a set of errors, so this is a bad API.

For the second problem, it would be much more sane to return a multierr
or a list of errors. If there's no fix for the first, I think it should
be removed from the lib.
2016-12-04 21:06:08 -05:00
James Shubin
639afe881c resources: Reduce logging on Send/Recv
This was too noisy, let's tone it down a bit.
2016-12-03 01:44:36 -05:00
James Shubin
2e718c0e9d resources: Improve notification system and notify refreshes
Resources can send "refresh" notifications along edges. These messages
are sent whenever the upstream (initiating vertex) changes state. When
the changed state propagates downstream, it will be paired with a
refresh flag which can be queried in the CheckApply method of that
resource.

Future work will include a stateful refresh tracking mechanism so that
if a refresh event is generated and not consumed, it will be saved
across an interrupt (shutdown) or a crash so that it can be re-applied
on the subsequent run. This is important because the unapplied refresh
is a form of hysteresis which needs to be tracked and remembered or we
won't be able to determine that the state is wrong!

Still to do:
* Update the autogrouping code to handle the edge notify properties!
* Actually finish the stateful bool code
2016-12-03 01:35:31 -05:00
James Shubin
b0a8fc165c resources: Improve the state/cache system
Refactor the state cache into the engine. This makes resource writing
less error prone, and paves the way for better notifications.
2016-12-03 00:07:29 -05:00
James Shubin
ba6044e9e8 resources, pgraph: split logical chunks into separate files 2016-12-03 00:07:29 -05:00
James Shubin
7f1c13a576 resources: Implement Send -> Recv
This is a new design idea which I had. Whether it stays around or not is
up for debate. For now it's a rough POC.

The idea is that any resource can _produce_ data, and any resource can
_consume_ data. This is what we call send and recv. By linking the two
together, data can be passed directly between resources, which will
maximize code re-use, and allow for some interesting logical graphs.

For example, you might have an HTTP resource which puts its output in a
particular file. This avoids having to overload the HTTP resource with
all of the special behaviours of the File resource.

For our POC, I implemented a `password` resource which generates a
random string which can then be passed to a receiver such as a file. At
this point the password resource isn't recommended for sensitive
applications because it caches the password as plain text.

Still to do:
* Statically check all of the type matching before we run the graph
* Verify that our autogrouping works correctly around this feature
* Verify that appropriate edges exist between send->recv pairs
* Label the password as generated instead of storing the plain text
* Consider moving password logic from Init() to CheckApply()
* Consider combining multiple send values (list?) into a single receiver
* Consider intermediary transformation nodes for value combining
2016-12-03 00:07:29 -05:00
James Shubin
63c5e35e2b misc: Cleanup unnecessary use of %v and small nitpicks 2016-12-03 00:07:29 -05:00
James Shubin
62e6a7d7fa resources: Add VarDir support
This gives resources a private directory where they can store state.
2016-12-03 00:07:29 -05:00
James Shubin
e5a3dae332 misc: Exclude vendor/ directory from ack matching
This is almost always what we want.
2016-12-03 00:07:29 -05:00
James Shubin
b45a7663b3 readme: Add video from NLUUG
This video shows the virt+cockpit PoC & the live remote execution demo.
2016-12-03 00:06:03 -05:00
Vinzenz Feenstra
6ef904f62b misc: Prefer dnf over yum when present
Signed-off-by: Vinzenz Feenstra <vfeenstr@redhat.com>
2016-11-22 12:26:17 +01:00
James Shubin
6d21cf3084 readme: Add video link from High Load Strategy 2016-11-18 16:16:51 -05:00
James Shubin
32bd96b6e2 resources: nspawn: Update grammer for Joe
Joe says this is the correct grammar. I incorrectly changed it wrongly.
2016-11-18 16:15:21 -05:00
James Shubin
fb5da76247 resources: Add stable go-systemd branch for now
We'll update this to point to upstream or JoeJulian once it stops
randomly changing and breaking git master!
2016-11-14 20:03:19 -05:00
James Shubin
e588f51824 resources: nspawn: Use new API
We broke the API (*cough* Joe), this updates it with the new version.
2016-11-14 19:36:13 -05:00
Joe Julian
3e419c4955 nspawn: bump go-systemd commit
joejulian/go-systemd was updated to provide more machine1 features and
was rebased to the latest master
2016-11-14 11:45:07 -08:00
James Shubin
606d2bafac resources: nspawn: Tweaks and updates
Here are some small fixes to enhance the original nspawn patch.
2016-11-12 00:43:55 -05:00
Joe Julian
8ac3c49286 nspawn: Add systemd-machined support for nspawn containers
This adds a rudimentary resource for systemd-machined's nspawn
containers, ensuring they're either started or stopped.
2016-11-11 14:55:14 -08:00
James Shubin
534aa84ed0 etcd: Watch for obvious failures on first startup
We should probably wait for this signal elsewhere too.
2016-11-11 07:37:36 -05:00
Vinzenz Feenstra
04d17cb580 examples: rename hostname.yml to hostname.yaml
Signed-off-by: Vinzenz Feenstra <vfeenstr@redhat.com>
2016-11-11 12:51:55 +01:00
Vinzenz Feenstra
d039006eb4 resources: Add new hostname resource
This resource allows to set and watch the hostname on a system.

Signed-off-by: Vinzenz Feenstra <vfeenstr@redhat.com>
2016-11-11 12:42:04 +01:00
James Shubin
fb04f62115 resources: file: Allow undefined file contents
An undefined file contents means we aren't managing that state!
2016-11-10 06:06:41 -05:00
James Shubin
3bffccc48e resources: Clean up errors and string printing 2016-11-08 03:49:27 -05:00
Vinzenz Feenstra
eef9abf0bf virt: Authentication support
Signed-off-by: Vinzenz Feenstra <vfeenstr@redhat.com>
2016-11-07 13:28:30 +01:00
Vinzenz Feenstra
de5ada30b7 virt: Avoid parsing URI all the time
Signed-off-by: Vinzenz Feenstra <vfeenstr@redhat.com>
2016-11-07 13:22:40 +01:00
Joe Julian
12f7d0a516 virt: Add logic to parse the libvirt uri
Guess the domain and os types from the libvirt uri and add the ability
to set the init for lxc containers.
2016-11-07 13:22:40 +01:00
Joe Julian
0aa9c7c592 Add IntelliJ Idea to .gitignore 2016-11-07 13:22:40 +01:00
Joe Julian
2216c8dc1c virt: Do not restrict VIR_ERR_NO_DOMAIN to qemu.
Consider it sufficent that a libvirt.VIR_ERR_NO_DOMAIN was reported
without concerning ourselves over who reported it.
2016-11-07 13:22:40 +01:00
James Shubin
984270ebe1 pgraph: Fix ineffassign warning
This ineffassign didn't seem to cause problems, but perhaps we didn't
exhaustively test all the areas.

Watch out to make sure this doesn't break any of the sync requirements,
eg: "A WaitGroup must not be copied after first use."
https://golang.org/pkg/sync/#WaitGroup
2016-11-04 02:47:13 -04:00
James Shubin
2e2658ab6f examples: make the libmgmt example more fun
You can try it out yourself by running `go build` and then calling it.
Use a bare integer argument to create that number of noop resources.
There are clearly some performance optimizations that we could do for
extremely large graphs.
2016-11-03 04:18:26 -04:00
James Shubin
1370f2a76b gapi: Split out graph generation into a proper graph API
This is a monster patch that splits out the yaml and puppet based graph
generation and pushes them behind a common API. In addition alternate
pluggable GAPI's can be easily added! The important side benefit is that
you can now write a custom GAPI for embedding mgmt!

This also includes some slight clean ups that I didn't find it worth
splitting into separate patches.
2016-11-03 03:56:16 -04:00
Joe Julian
75dedf391a virt: don't set emulator path
Remove the implicit emulator path from the domain definition. Libvirt is
already configured to use the correct emulator for kvm or qemu and
specifying it creates distro dependence.

Fixes #85
2016-10-27 15:20:06 -04:00
Juergen Hoetzel
7b5c640d05 readme: Fix go get command
"go get" requires a package name
2016-10-27 18:30:00 +02:00
James Shubin
aa9a21b4d0 cli: Pass through program and version strings
We forgot to pass these through. If they're undefined, it errors.
2016-10-24 17:41:03 -04:00
James Shubin
71de8014d5 main: Libify mgmt with a golang API
This is an initial implementation of a possible golang API. In this
particular version, the *gconfig.GraphConfig data structures are
emitted, instead of possibly building a pgraph. As long as we can
represent any local graph as the data structure, then this is fine!

Is there a way to merge the gconfig Vertex and the pgraph Vertex?
2016-10-24 17:33:31 -04:00
Marc Antoine Dumont
80476d19f9 Add the link to a new dependencies in README.md
Add the link to the dependencise github.com/rgbkrk/libvirt-go
2016-10-24 16:16:05 -04:00
James Shubin
15103d18ef readme: Update README file to make it clearer for new hackers 2016-10-23 20:47:12 -04:00
James Shubin
0dbd2004ad main: Split apart logic in main
This splits most of the main logic from the cli logic so that they can
be used independently, in particular for if we ever libify mgmt.
2016-10-23 20:23:04 -04:00
James Shubin
8c92566889 resources: virt: Update CPUs variable to new uint16 signature
Now things are consistent after my new patch upstream!
2016-10-23 02:41:32 -04:00
James Shubin
fb9449038b resources: Update constructor signature to return error as well
Update the helper functions so they're easier to properly use!
2016-10-23 01:36:34 -04:00
James Shubin
e06c4a873d resources: Set the defaults for metaparameters
This now lets us have defaults for metaparameters that aren't the zero
value for that type.
2016-10-23 01:14:02 -04:00
James Shubin
c4c28c6c82 spec: Improve the rpm package
This still needs a lot of work by a packaging specialist.
2016-10-19 20:10:11 -04:00
James Shubin
42ff9b803a resources: Use Events() method instead of raw channel
This makes things easier if we ever split resources out into separate
packages.
2016-10-19 20:08:53 -04:00
James Shubin
3831e9739c resources: virt: Update to new function signature
This changed in git master, and is now more idiomatic.
2016-10-19 13:59:20 -04:00
James Shubin
f196e5cca2 test: Fix travis so it pulls in our deps 2016-10-19 13:51:38 -04:00
Joe Julian
d3af9105ee Use the download-only flag when fetching dependencies 2016-10-19 09:54:15 -07:00
James Shubin
6d685ae4d6 misc: Add libvirt header file dependency 2016-10-19 04:23:15 -04:00
James Shubin
8381d8246a resources: virt: Add a virt resource based on libvirt
This adds an initial implementation of a virt resource based on libvirt.
It is not complete and requires more testing. The initial skeleton was
written by nseps but was not merged. It was later cleaned up and merged
in its current form by purpleidea. Many thanks to nseps for getting this
going, and hopefully we'll get you contributing more in the future!
2016-10-19 04:11:17 -04:00
James Shubin
b26322fc20 all: Rename UUID to UID.
Felix pointed out that these ID's are unique, but not universally unique
across the cluster, which might be confusing to new programmers. As a
result, rename them all.
2016-10-18 23:03:55 -04:00
James Shubin
1c1e8127d8 resources: Check that resource kind is set.
This could be a Fatal instead, but might as well fail gracefully.
2016-10-18 14:07:27 -04:00
James Shubin
1b3b4406ff resources: Interfaces parameters can be named to help documentation 2016-10-18 14:04:47 -04:00
James Shubin
cf0b77518a resources: List resources in alphabetical order 2016-10-18 14:04:21 -04:00
James Shubin
afdbf44e23 make: Sed needs g option to replace multiple PROGRAM names 2016-10-13 10:23:18 -04:00
Alexandre-Xavier Labonté-Lamoureux
ec87781956 test: Tokens should always have a colon 2016-10-11 13:46:59 -04:00
James Shubin
a6ae958be7 etcd: Fix type issue
I was lazy and pushed the previous fix too quickly. Sorry, fixed now!
2016-10-07 15:59:53 -04:00
James Shubin
312103ef1b test: update lint checker to support packages 2016-10-07 15:51:58 -04:00
James Shubin
c2911bb2b7 etcd: Verify struct is not nil before accessing retries value
This didn't happen often because there's a nominateCallback race, but is
a bug which happened occasionally.
2016-10-07 15:36:09 -04:00
James Shubin
8ca5e38121 readme: Update repository with information about remote execution 2016-10-07 15:35:29 -04:00
James Shubin
4b8ad3a8a7 godoc: Document packagekit package 2016-10-03 15:28:25 -04:00
James Shubin
f219c2649d godoc: Document resources package 2016-10-03 15:26:41 -04:00
James Shubin
cfde54261b golint: Outdent else statement 2016-10-03 15:21:08 -04:00
James Shubin
71a82b0a34 resources: Use named fields to avoid bothering go vet
This removes an invalid warning which tries to prevent ambiguous changes
I suppose.
2016-10-03 15:02:41 -04:00
James Shubin
b7bd2d2664 recwatch: Rearrange condition to unconfuse golinter
This wasn't a bug, but the go linter is unhappy because it is dumb.
2016-10-03 14:58:08 -04:00
James Shubin
cd26a0770d test: Catch go vet issues in subpackages
Improve our tests now that we have multiple packages to test.
2016-10-03 14:50:28 -04:00
James Shubin
46893e84c3 recwatch: Split recursive watching into sub-package
This splits the recursive watching bit of the file file resource into
it's own package. This also de-duplicates the configwatch code and puts
it into the same package. With these bits refactored, it was also easier
to clean up the error code in main.
2016-10-03 14:48:57 -04:00
James Shubin
567dcaf79d resources: fix deadlock on recursion
This was a sneaky deadlock which wasn't 100% reproducible. It was pretty
tricky to find because the same deadlocked behaviour was seen due to the
regression in the fsnotify module.

The event system needs a bit of a cleaning, but this can happen later.
2016-10-01 08:19:46 -04:00
James Shubin
9368c7e05f resources: msg: Turn on journal logging in the example
Make this more useful by default.
2016-09-28 05:59:55 -04:00
James Shubin
654b3e9dbe file: Fix regression in fsnotify code
This was a major deadlock that hit the file resource. I didn't notice it
earlier because I was using an older version of fsnotify and I hadn't
done a go get -u to refresh it. I finally tracked this down, and will
vendor the repository until a fix upstream or a workaround downstream is
added.

The upstream issue is: https://github.com/fsnotify/fsnotify/issues/123
2016-09-28 03:26:33 -04:00
James Shubin
f09db490f0 golint: Fix nested else statement 2016-09-27 13:33:58 -04:00
Felix Frank
30d93cfde7 resources: msg: Introduce new resource type to log arbitrary messages
Untested things:
* systemd journal

Unimplemented things:
* syslog
2016-09-27 13:31:16 -04:00
James Shubin
41b3db7d6b golint: Fix issues caught by the linter 2016-09-27 12:14:12 -04:00
James Shubin
2a60debceb resources: svc: Catch reload states too 2016-09-27 11:47:55 -04:00
James Shubin
eb30642b6f readme: Add link to godoc 2016-09-27 08:54:02 -04:00
James Shubin
ea85e2af6b golang: Update to version 1.6 as the minimum
Etcd now requires golang version 1.6 or greater.
2016-09-27 08:08:01 -04:00
James Shubin
ef979a0839 vendor: Update etcd to working version and fix grpc stuff 2016-09-27 07:57:04 -04:00
Felix Frank
e0107b1dda have 'make clean' get rid of generated code files again 2016-09-27 05:20:56 -04:00
James Shubin
ccc00f913d Resources: Update interface to support errors
This was an early interface mistake I made and is now corrected.
We'll plumb in checking the error state of Init() and running Validate()
later on.
2016-09-27 05:16:37 -04:00
James Shubin
ad3c6bdc88 TODO: add TODO item about DHT idea 2016-09-27 05:07:34 -04:00
James Shubin
8fe3891ea9 Makefile: Limit depth for yamlfmt 2016-09-27 05:06:43 -04:00
James Shubin
63f21952f4 golang: Split things into packages
This makes this logically more separate! :) As an aside...

I really hate the way golang does dependencies and packages. Yes, some
people insist on nesting their code deep into a $GOPATH, which is fine
if you're a google dev and are forced to work this way, but annoying for
the rest of the world. Your code shouldn't need a git commit to switch
to a a different vcs host! Gah I hate this so much.
2016-09-26 12:30:28 -04:00
James Shubin
361d643ce7 resources: Compare Delay and Retry metaparams on graph change
We forgot to add a number of important properties to the Compare. This
means that we'll rebuild elements for these changes too now.
2016-09-20 05:35:04 -04:00
Antoine Racine
abe1ffaab6 Add pcap.h dependency on Debian/Ubuntu to fix make deps error 2016-09-19 22:51:10 -04:00
James Shubin
fc24c91dde Resources: Add retry and retry delay meta parameters
All resources can now set a retry limit (-1 for infinite) and a delay
between retries. This applies to both the CheckApply methods, and the
Watch methods as well. They each have their own separate counts, but use
the same input meta param, since I decided it wouldn't be useful to have
a separate watchRetry and watchDelay set of meta parameters.

In the process, we got rid of about 15 error cases which would normally
panic.

This patch required a slight overhaul of the Event system.

The previous commit is an earlier version of this patch which I decided
to leave in to "show my work" as I used to have to do in math class.
It's slightly more correct with the current event system, and this
version is less correct and has a few bugs, but that is because the
event system needs a massive overhaul, and once that's done this should
all work properly for the corner cases.
2016-09-19 06:32:21 -04:00
James Shubin
53cabd5ee4 Resources: Prototype retry and retry delay meta parameters
This was the initial cut of the retry and delay meta parameters.
Instead, I decided to move the delay action into the common space
outside of the Watch resource. This is more complicated in the short
term, but will be more beneficial in the long run as each resource won't
have to implement this part itself (even if it uses boiler plate).

This is the first version of this patch without this fix. I decided to
include it because I think it has more correct event processing.
2016-09-19 06:32:21 -04:00
James Shubin
2b1e8cdbee travis: bump to newer golang version (fedora 23) 2016-09-19 03:39:09 -04:00
Michaël Faille
9715146495 travis : use go1.7 to be future proof.
And, to reflect my own config `;-)`
2016-09-19 03:37:55 -04:00
Michaël Faille
22b0b89949 Let env find bash in shebang.
On Nixos and GNUIX-SD, bash is chroot in package store path.
In my case : /nix/store/qvccmr6fsis4kqlvlk8pb1c8c0r0cwai-system-path/bin/bash

In any case, using `/usr/bin/env bash` is the recommended way to get bash
portable across UNIX-like systems.
2016-09-19 03:36:38 -04:00
Michaël Faille
2ebc23a777 cli : Prefer github.com/urfave/cli as cli
Over  github.com/codegangsta/cli
Note : I just adapt the README.md and dependency manager
2016-09-19 02:26:28 -04:00
James Shubin
0199285319 readme: Add information for new users 2016-09-18 05:59:46 -04:00
James Shubin
277ab2fe44 readme: We now have an arch aur build.
Thanks Joe Julian!
2016-09-18 05:52:48 -04:00
James Shubin
8a96dfdc8a art: Add a logo
Sarah Jane Cox made us a logo. Thanks SJ!
2016-09-18 05:47:33 -04:00
James Shubin
66fbbb940a test: temporarily disable test 2016-09-18 03:46:42 -04:00
James Shubin
716ea1bb3c etcd: Update examples to be more useful for single machine tests 2016-09-14 23:00:59 -04:00
James Shubin
3d701d3daa todo: Update TODO file 2016-09-12 02:01:44 -04:00
James Shubin
598c74657c file: Overhaul file resource and add recursion
The file resource contained some of the early golang code that I wrote
for this project. Needless to say, some of it was quite yucky, and it
was also lacking a number of important features. This patch builds upon
it so that it starts being usable for directories of files too.

Many thanks to Sam Gélineau for helping with the recursive watching. My
brain officially didn't want to look at that code anymore.
2016-09-12 01:55:31 -04:00
Felix Frank
4bd53d5ab0 add puppet support documentation 2016-09-12 02:40:09 +02:00
James Shubin
70f8d54a31 thanks: More people really needed to be on the list 2016-09-07 19:41:39 -04:00
James Shubin
4ef25a33fc docs: Add an FAQ entry about difference between two similar methods
Felix wins the points for first asking the question I knew would
eventually come but didn't document earlier.
2016-09-07 19:31:13 -04:00
Felix Frank
f5dd90a8dd add unit tests for UUID comparison and resource event passing 2016-09-08 01:12:56 +02:00
James Shubin
a84defd689 readme: Add new blog posts by Felix
The last post here was one of my favourites! Super impressive stuff!
2016-09-07 01:08:21 -04:00
James Shubin
1cf88d9540 test: Increase timeouts of t8
Increase the timeouts in the rare chance that this is slow performing
travis, and not just an etcd regression.
2016-09-02 02:27:39 -04:00
James Shubin
644a0ee8c8 puppet: golint fixes 2016-09-02 02:25:41 -04:00
James Shubin
e9d5dc8fee packagekit: golint fixes 2016-09-02 02:23:13 -04:00
James Shubin
8003202beb resources: golint fixes 2016-09-02 02:03:26 -04:00
James Shubin
b46432b5b6 misc: Golint fixes 2016-09-02 01:46:45 -04:00
James Shubin
5e3f03df06 etcd: Catch possible raft grpc error 2016-09-02 00:35:05 -04:00
James Shubin
8ab8e6679a test: provider usage text for shell test runner 2016-09-01 22:52:32 -04:00
James Shubin
786b896018 remote: small cleanups to update misc notes 2016-08-31 22:56:00 -04:00
James Shubin
40723f8705 docs: Add additional documentation about remote execution 2016-08-31 22:42:09 -04:00
James Shubin
2a0721bddf remote: allow converge during corner cases
This allows the system to converge during corner cases where there is an
error, or when there are no remotes being used, but we are using the
--no-watch variable.

I deliberately left this in as a separate commit instead of rebasing
into the remote execution development branch because the placement of
the Unregister() and semaphore.V(1) were quite subtle and easy to forget
about.
2016-08-31 22:42:09 -04:00
James Shubin
ff01e4a5e7 remote: Add distributed converged timeout
This patch extends the --converged-timeout argument so that when used
with --remote it waits for the entire set of remote mgmt agents to
converge simultaneously before exiting.

purpleidea says: This particular part of the patch probably took as much
work as all of the work required for the initial remote patches alone!
2016-08-31 21:55:19 -04:00
James Shubin
6794aff77c miscellaneous cleanups and fixes 2016-08-31 21:55:19 -04:00
James Shubin
636f2a36b1 etcd: Use new converged timers and allow skipping them
This implements the new extensions to the converged UUID API so that we
can keep a consistent timer running and reset it when needed. This is
useful because we now allow certain Watcher callbacks and other
operations to explicitly _not_ cause a convergerUUID to un-converge.
This is a necessary dependency for the distributed remote
converged-timeout work.
2016-08-31 21:55:19 -04:00
James Shubin
eee652cefe converger: Add new timer system for determining convergence
This adds a new method of marking whether a particular UUID has
converged or not. You can now Start, Stop, or Reset a convergence timer
on the individual UUID's. This wraps the existing SetConverged calls
with a hidden go routine. It is not recommended to use the SetConverged
calls and the Timer calls on the same UUID.
2016-08-31 21:55:19 -04:00
James Shubin
6d45cd45d1 converger: Add new methods to the API
This adds new helper methods to the API such as the ability to query the
set timeout, and to set the state change function after initialization.
The first makes it easier to pass in the timeout value to functions and
structs, because you don't need to pass it separately from the converger
object. The second makes it easy to specify the state change functon
when you don't know what it is at creation time. In addition, it is more
powerful now, and tells you when we converge or un-converge in case we
want to take different actions on each.
2016-08-31 21:55:19 -04:00
James Shubin
f5fb135793 converger: Update the API for errors and naming
We updated the API and behaviour to make two changes:

1) We remove the log.* stuff, and replace the Fatal calls with straight
calls to panic(). These are meant to be like an assert, and shouldn't
happen unless there is a user error or a bug.

2) We made the !uuid.IsValid() checks return an error instead of causing
a panic. These returns errors instead, and makes the process safer for
users who are okay calling a function that won't have an effect.
2016-08-31 21:55:19 -04:00
James Shubin
6bf32c978a etcd: Rename loop to be more consistent in messages
Small nitpick fixups
2016-08-31 21:55:19 -04:00
James Shubin
8d3011fb9c etcd: Add a timeout for etcd server to start correctly
This also updates etcd to a newer version with a fix that allows this
detection and timeout operation to be possible.
2016-08-31 21:55:19 -04:00
James Shubin
9260066fa3 tests: Workaround regression in two host etcd clusters
If you don't give your two host cluster enough time to "feel healthy",
it will generate an error if you do operations within five seconds. This
is a regression and the five seconds is also quite arbitrary. This is
detailed at: https://github.com/coreos/etcd/issues/6305

This seems to be a bit of a race condition, even with a 10s timer, so
this also disables the StrictReconfigCheck. Re-enable this as soon as
possible.
2016-08-31 21:55:19 -04:00
James Shubin
5e45c5805b Improve internal etcd error handling 2016-08-31 21:55:19 -04:00
James Shubin
db4de12767 Add more flexibility around the prefixes available
This allows you to specify a custom prefix, or a tmp prefix which is
chosen automatically.
2016-08-31 21:55:19 -04:00
James Shubin
d429795737 Improve the configWatcher array to allow N files
This simplifies the code in main and hides it in the watcher!
2016-08-31 21:55:19 -04:00
James Shubin
276219a691 Run tests with a tmp prefix
This will avoid failures if /var/lib/mgmt/ isn't writable.
2016-08-31 21:55:19 -04:00
James Shubin
03c1df98f4 Improve prefix creation and feedback
This makes the default /var/lib/mgmt/ directory, but also allows you to
use a prefix in /tmp/ automatically if you can't write anywhere else.
2016-08-31 21:55:19 -04:00
James Shubin
79ba750dd5 Automatically update remote files on change
This extends the automatic watching of graph definition files across the
remote SSH boundary.
2016-08-31 21:55:19 -04:00
James Shubin
1d0e187838 Etcd: switch to using a directory prefix 2016-08-31 21:55:19 -04:00
James Shubin
ad1e48aa2d Add caching for remote execution
This speeds up copying of the binary for slow connections. It also
finally adds a universal directory prefix for mgmt!
2016-08-31 21:55:19 -04:00
James Shubin
7032eea045 Remote "agent-less" mode
This is a new mode to be used for bootstrapping mgmt clusters or in
situations with tight operational restrictions.

This includes the basics, additional functionality will follow!
2016-08-31 21:55:19 -04:00
James Shubin
bdb970203c Start converger even if graph is empty 2016-08-30 17:47:25 -04:00
James Shubin
fa4f5abc78 Fix up tests, and one small bug 2016-08-30 17:47:25 -04:00
James Shubin
0c7b05b233 Check for exit status of tests 2016-08-30 17:47:25 -04:00
Joe Julian
4ca98b5f17 Allow make overrides of version if git fails
If the $(shell ...) command fails, the ':=' operator fails as well
preventing variable overrides from functioning.

Wrap the assignment in an $(or ...) to prevent the shell script from
running if the variable is set from the command line or the environment.

Fixes issue #58
2016-08-30 14:44:03 -07:00
Joe Julian
4e00c78410 Change "uname -i" to "uname -m" to be portable
According to the documentation for uname:

       -m, --machine
              print the machine hardware name

       -i, --hardware-platform
              print the hardware platform (non-portable)

So use the portable -m version.
2016-08-30 11:55:45 -07:00
James Shubin
17adb19c0d Fix typo 2016-08-04 00:44:50 -04:00
James Shubin
1db936e253 Update docs because they were out of date 2016-08-03 05:28:18 -04:00
James Shubin
7194ba7e0e Update docs to add automatic clustering 2016-08-03 05:16:47 -04:00
James Shubin
59b9b6f091 Docs: Add FAQ entry about similarly named band 2016-08-02 04:29:56 -04:00
James Shubin
c1ec8d15f3 Improve README for first time users
We need --recursive because we have a dependency vendored this way.
2016-08-02 04:25:35 -04:00
James Shubin
24ba6abc6b Don't block on member exits
The MemberList() function which apparently looks at all of the endpoints
was blocking right after a member exited, because we still had a stale
list of endpoints, and while a member that exited safely would update
the endpoints lists, the other member would (unavoidably) race to get
that message, and so it might call the MemberList() with the now stale
endpoint list. This way we invalidate an endpoint we know to be gone
immediately.

This also adds a simple test case to catch this scenario.
2016-07-26 04:23:46 -04:00
James Shubin
f6c1bba3b6 Avoid a rare panic if DestroyServer is called early
I never actually hit this bug, but I noticed it was possible when
examining the WaitGroup code that gets .Done() by DestroyServer().
2016-07-26 01:58:13 -04:00
James Shubin
a606961a22 Be safe when closing in destroy in case client is nil 2016-07-25 21:46:08 -04:00
James Shubin
cafe0e4ec2 Round of golint fixes to improve documentation.
This is really boring :(
2016-07-25 21:36:09 -04:00
James Shubin
e28c1266cf Do some gofmt simplifications 2016-07-25 20:56:33 -04:00
James Shubin
c1605a4f22 Add test case for urfave regression
Credit to jerith for helping me hack this together :)
2016-07-25 20:25:08 -04:00
James Shubin
7aeb55de70 Port embedded etcd to embed API
This also updates the vendored version of etcd to current git master,
which is the only place this is supported at the moment.
2016-07-25 19:10:09 -04:00
James Shubin
8ca65f9fda Revert "Copy in out of tree patches"
This reverts commit d26b503dca.

Use new etcd "embed" API.
2016-07-24 00:08:58 -04:00
James Shubin
94524d1156 Revert "Revert "Allow 1.6 to fail for now""
This reverts commit 78d769797f.
2016-07-20 03:09:17 -04:00
Sharad Ganapathy
a1ed03478b Adding timer resource and usage examples 2016-07-17 14:01:36 -04:00
James Shubin
402a6379b9 Add exec3 example
This is meant to be easier to understand than just sleep's.
2016-07-14 17:37:36 -04:00
Jeremy Thurgood
5d45bcd552 Check for old golang versions while installing dependencies. 2016-07-07 10:14:56 +02:00
Jack Henschel
f1fa64c170 Add recording and slides from DebConf16 presentation by James Shubin 2016-07-06 08:40:30 +02:00
Jack Henschel
50fc78564c Remove noop functionality from TODO
noop functionality (Bug #21) has been implemented:
6bbce039aa
9f56e4a582
2016-07-05 21:33:59 +02:00
James Shubin
3e5863dc8a Get travis results faster!
Thanks to jerith for the tip
2016-07-05 07:07:44 -04:00
James Shubin
94b447a9c5 Remove manual etcd usage. No longer needed. 2016-07-05 06:58:04 -04:00
James Shubin
78d769797f Revert "Allow 1.6 to fail for now"
This reverts commit f63b1cd56d.

This now works with stable version of etcd. Let's hope it stays that
way! :)
2016-07-05 04:32:38 -04:00
James Shubin
672baae126 Bump to etcd v3.0.0 2016-07-05 04:21:43 -04:00
James Shubin
e942d71ed2 Include video 2016-06-20 12:57:51 -04:00
James Shubin
f5d24cf86c New blog post! 2016-06-20 12:30:37 -04:00
James Shubin
f63b1cd56d Allow 1.6 to fail for now
I'm going to let 1.6 fail for now until F24 comes out and I can start
testing on 1.6 -- I'd like to be able to test all versions, but I don't
have all the resources to do so right now. If you want to help, please
let me know!
2016-06-20 00:43:54 -04:00
James Shubin
66719b3cda Update links 2016-06-20 00:36:46 -04:00
James Shubin
a5e9f6a6fc Fix stupid gofmt issue 2016-06-19 02:47:36 -04:00
James Shubin
f821afdf3e Update cli library path 2016-06-19 02:35:45 -04:00
James Shubin
2c61de83c6 Add go report card! 2016-06-19 02:35:24 -04:00
James Shubin
6da6f75b88 Add vendored etcd
This is a git submodule. Once etcd v3 becomes stable, this might not be
necessary anymore. We'll have to wait and see!
2016-06-18 04:43:19 -04:00
James Shubin
a55807a708 Split formatting into two targets 2016-06-18 04:43:19 -04:00
James Shubin
fce86b0d08 docs: add faq entry about using external etcd cluster 2016-06-18 04:43:19 -04:00
James Shubin
d26b503dca Copy in out of tree patches
These patches are proposed upstream changes and code for and from etcd.
Ideally we would revert this patch when/if things are merged upstream!
The majority of the work is in: https://github.com/coreos/etcd/pull/5584
2016-06-18 04:43:19 -04:00
James Shubin
5363839ac8 Embedded etcd
This monster patch embeds the etcd server. It took a good deal of
iterative work to tweak small details, and survived a rewrite from the
initial etcd v2 API implementation to the beta version of v3.

It has a notable race, and is missing some features, but it is ready for
git master and external developer consumption.
2016-06-18 04:43:19 -04:00
James Shubin
715a4bf393 Update links 2016-06-18 00:35:16 -04:00
Felix Frank
8f83ecee65 add puppet integration code
Puppet can be used on the basis of the ffrank-mgmtgraph module.
There are three modes available:
* fetching catalogs from the master (--puppet agent)
* compiling a manifest from a local file (--puppet /path/to/file.pp)
* compiling a manifest from the cli (--puppet "<manifest>")

Catalogs from the master are currently never refreshed. We should
add some more code to re-run the parsing function at an interval
equal to Puppet's local 'runinterval' setting.

There is also still a distinct lack of tests.

Still, this fixes #8
2016-06-01 00:34:40 +02:00
Felix Frank
2eed4bda42 README: add the capnslog package to the build dependencies 2016-06-01 00:32:36 +02:00
Raphaël Pinson
f4e1e24ca7 Typo in DOCUMENTATION.md 2016-05-26 10:50:51 +02:00
Paul Morgan
05c540e6cc move doc files to /usr/share/doc/mgmt and tag as docs
After this commit:

* doc files are in `/usr/share/doc/mgmt-*/`
* `rpm -qd mgmt` lists the files
* docs are listed on single lines in spec file
  to minimize future diff churn
* docs are in alpha order
2016-05-24 02:23:43 +00:00
James Shubin
9656390c87 Add quote from notable automation specialist and author 2016-05-22 02:16:19 -04:00
James Shubin
4b6470d1e1 Remove pesky spaces 2016-05-21 12:57:28 -04:00
Michał Czeraszkiewicz
56471c2fe4 Add Docker support 2016-05-21 11:08:14 +02:00
James Shubin
9f56e4a582 Add global --noop support
This is part two of the earlier patch in
6bbce039aa

We also rename GetMeta to just Meta to clean up the API.
2016-05-18 14:28:34 -04:00
James Shubin
12ea860eba Expand the event system slightly
This also adds a cleaner exit for the inner main loop. I'm not sure if
it's absolutely needed, but this will give me more confidence that we
won't end in the middle of some action.
2016-05-18 11:57:36 -04:00
James Shubin
b876c29862 Add logging workaround when embedding etcd
This was discussed in: https://github.com/coreos/etcd/issues/4115
2016-05-18 11:56:51 -04:00
Martin Alfke
6bbce039aa noop as resource meta param
first part of #21
tested with example/noop1.yaml on CentOS 6
2016-05-17 11:43:30 -04:00
Martin Alfke
1584f20220 make go vet installation optional 2016-05-17 16:47:52 +02:00
Martin Alfke
dcad5abc1c remove vet
go 1.6 has vet included
https://github.com/hashicorp/vault/issues/1310#issuecomment-207922338
2016-05-17 16:47:33 +02:00
James Shubin
ab73261fd4 Fix cli API change
I guess this is why builds were breaking. Remember to go get -u ...
2016-05-14 16:26:09 -04:00
James Shubin
05b75c0a44 Pkg: Immediately unconverge on events
It's more correct to run this as soon as possible. This wasn't because
of any known bug, but it's more correct stylistically.
2016-05-14 15:27:01 -04:00
James Shubin
ba7ef0788e Pkg: cache state when it is correct
We forgot to cache the state when we are converged. This avoids
repetitive checking when we hit repeated backpoke()'s for example.
2016-05-14 12:28:42 -04:00
Felix Frank
3aaa80974e rename the 'stateok' return value to 'checkok'
The naming was confusing because the boolean return value expresses
whether the resource needed changing (the check failed) as opposed to
the state not being not OK.

purpleidea note: The "stateok" (now properly renamed to "checkok") is
actually the historical bool return value of the Check() -> bool
function which is now part of the CheckApply() amalgamation. This is an
easy way to think about it if you're trying to understand why at the end
of a successful apply we return false, nil.
2016-05-14 18:15:06 +02:00
Felix Frank
995ca32eee packagekit: support Debian's 'all' architecture 2016-05-14 18:15:06 +02:00
Martin Alfke
bf5f48b85b find on OS X needs the dot 2016-05-13 14:30:23 +02:00
James Shubin
d6e386a555 Add a base method for GroupCmp
You'll want to override this when implementing a resource that does
grouping.
2016-04-28 15:37:19 -04:00
James Shubin
a0a71f683c Fix typo
Exec doesn't group, so this never impacted anything, but wasn't correct.
2016-04-28 15:35:32 -04:00
James Shubin
7adf88b55b Update TODO file 2016-04-26 01:15:59 -04:00
Paul Morgan
8a9d47fc4b move systemd tip from README to DOCUMENTATION
This consolidates how-to at the right place for the docs.
2016-04-26 03:07:59 +00:00
James Shubin
2a0a69c917 Update Makefile to be more useful for hackers 2016-04-25 22:52:08 -04:00
James Shubin
aeab8f55bd Fix go generate issue with path
Gah, I hate you $GOPATH...
2016-04-25 22:52:08 -04:00
James Shubin
9407050598 Add flatten helper to take apart messy nested lists
This isn't 100% necessary, but it's a friendly feature to add, and it
was a fun function to write.
2016-04-25 22:52:08 -04:00
Paul Morgan
b99da63306 add myself as a contributor 2016-04-26 02:19:24 +00:00
Paul Morgan
f0d6cfaae4 add systemd unit file
Update the spec file with the rpm macro to put the unit file
in the system-wide unit file directory based on:

    [root@1713bbf19a0b /]# rpmbuild --eval '%{_unitdir}'
    /usr/lib/systemd/system

Allow user to create a drop directory to specify options
via environment variables.

Resolves https://github.com/purpleidea/mgmt/issues/12.
2016-04-26 02:19:24 +00:00
Paul Morgan
3120628d8a allow to specify CLI options via environment vars
Use `git show -w` to inspect this commit diff since
it changes whitespace due to `gofmt`.

This commit allows to use environment variables in place of
CLI parameters to change program behavior. This supports the
notion of [12-factor apps](http://12factor.net/)
and makes it easier to dockerize the app as well as create
a systemd unit file.

It establishes a pattern of `MGMT_*` naming convention
for environment variables.
2016-04-26 02:19:24 +00:00
Paul Morgan
2654384461 add make target to build static binary
This is useful to generate a binary that can be dropped
onto any arbitrary distro, such as CoreOS, without having
to worry about glibc or other dependencies.

Specifically: CoreOS uses glibc, but it does not have a
package manager. It also has a read-only OS (`/usr/`).
Thus I'd like to compile a binary that can be dropped
into CoreOS and have zero dependencies.

* `make build` builds the same as it did before this commit.
* `make all` builds both dynamic and static bins, as expected.

I struggled with a way to DRY this up _and_ avoid diff churn.
In the end, I went with simplicity even though it's not DRY.
2016-04-26 02:17:20 +00:00
Paul Morgan
eac3b25dc9 improve portability of the Makefile
I like to build on Alpine Linux when possible, but
coreutils on Alpine does not include the
[`arch`](http://git.savannah.gnu.org/gitweb/?p=coreutils.git;a=blob_plain;f=src/coreutils-arch.c;hb=HEAD)
wrapper to print the kernel architecture.

`uname` and `arch` are both provided by coreutils rpm
on RedHat derivatives, so this change does not impact
the build on those distros.
2016-04-26 00:43:49 +00:00
Paul Morgan
7788f91dd5 make life easier for people who have EditorConfig plugin
If a person has the EditorConfig plugin,
this overrides default editor configs to play nicely with go files
and the Makefile.
2016-04-26 00:40:35 +00:00
James Shubin
d0c9b7170c Small nitpick fixups 2016-04-07 21:08:26 -04:00
Felix Frank
d84caa5528 don't abort test suite on first failure
The test.sh script aborts as soon as a test fails. This can save time
on the local machine, but is inconvenient in CI where an early minor
failure can mask more severe errors that will be found later.
2016-04-07 00:50:57 +02:00
James Shubin
2ab72bdf94 Allow stubborn users to avoid having to move their project
Some users like to have their project in ~/code/mgmt/ for example, but
this is not compatible with a $GOPATH which is elsewhere. This isn't a
problem unless you need vendored directories in ~/code/mgmt/vendor/
which doesn't work because ~/code/mgmt/ isn't in $GOPATH. With this
symlink and the provided ~/bin/go wrapper, vendored directories work
exactly as expected, and we also get a local $GOPATH pointing to the
same thing. Since $GOPATH must have a src/ dir, and vendor/ must NOT,
then we symlink the two together accordingly.

An important part of this is that those who like to put everything in
$GOPATH won't be affected by any of this!
2016-04-04 01:25:23 -04:00
James Shubin
f6833fde29 Update docs and README to mention new blog post
woo... it's done! thanks to my reviewers!
2016-03-30 06:08:31 -04:00
James Shubin
fa8a50b525 Update TODO file 2016-03-30 03:49:49 -04:00
James Shubin
d80c6bbf1d Add additional autogrouping example 2016-03-30 03:40:17 -04:00
James Shubin
6f3ac4bf2a Rework the converged detection and provide a clean interface
The old converged detection was hacked in code, instead of something
with a nice interface. This cleans it up, splits it into a separate
file, and removes a race condition that happened with the old code.

We also take the time to get rid of the ugly Set* methods and replace
them all with a single AssociateData method. This might be unnecessary
if we can pass in the Converger method at Resource construction.

Lastly, and most interesting, we suspend the individual timeout callers
when they've already converged, thus reducing unnecessary traffic, and
avoiding fast (eg: < 5 second) timers triggering more than once if they
stay converged!

A quick note on theory for any future readers... What happens if we have
--converged-timeout=0 ? Well, for this and any other positive value,
it's important to realize that deciding if something is converged is
actually a race between if the converged timer will fire and if some
random new event will get triggered. This is because there is nothing
that can actually predict if or when a new event will happen (eg the
user modifying a file). As a result, a race is always inherent, and
actually not a negative or "incorrect" algorithm.

A future improvement could be to add a global lock to each resource, and
to lock all resources when computing if we are converged or not. In
practice, this hasn't been necessary. The worst case scenario would be
(in theory, because this hasn't been tested) if an event happens
*during* the converged calculation, and starts running, the exit command
then runs, and the event finishes, but it doesn't get a chance to notify
some service to restart. A lock could probably fix this theoretical
case.
2016-03-29 20:27:38 -04:00
James Shubin
a6dc81a38e Allow failures in 1.4.x because of etcd problem
The error seen is:

req.Cancel undefined
(type *http.Request has no field or method Cancel)

And is possibly the result of something etcd folks changed. Sadly, the
fast building 1.4.x won't work at the moment :(
2016-03-28 21:28:08 -04:00
James Shubin
81c5ce40d4 Add grouping algorithm
This might not be fully correct, but it seems to be accurate so far. Of
particular note, the vertex order needs to be deterministic for this
algorithm, which isn't provided by a map, since golang intentionally
randomizes it. As a result, this also adds a sorted version of
GetVertices called GetVerticesSorted.
2016-03-28 21:16:03 -04:00
James Shubin
c59f45a37b Force process events to be synchronous
This avoids messing up the converged-timeout state!
2016-03-28 21:16:03 -04:00
James Shubin
1b01f908e3 Add resource auto grouping
Sorry for the size of this patch, I was busy hacking and plumbing away
and it got out of hand! I'm allowing this because there doesn't seem to
be anyone hacking away on parts of the code that this would break, since
the resource code is fairly stable in this change. In particular, it
revisits and refreshes some areas of the code that didn't see anything
new or innovative since the project first started. I've gotten rid of a
lot of cruft, and in particular cleaned up some things that I didn't
know how to do better before! Here's hoping I'll continue to learn and
have more to improve upon in the future! (Well let's not hope _too_ hard
though!)

The logical goal of this patch was to make logical grouping of resources
possible. For example, it might be more efficient to group three package
installations into a single transaction, instead of having to run three
separate transactions. This is because a package installation typically
has an initial one-time per run cost which shouldn't need to be
repeated.

Another future goal would be to group file resources sharing a common
base path under a common recursive fanotify watcher. Since this depends
on fanotify capabilities first, this hasn't been implemented yet, but
could be a useful method of reducing the number of separate watches
needed, since there is a finite limit.

It's worth mentioning that grouping resources typically _reduces_ the
parallel execution capability of a particular graph, but depending on
the cost/benefit tradeoff, this might be preferential. I'd submit it's
almost universally beneficial for pkg resources.

This monster patch includes:
* the autogroup feature
* the grouping interface
* a placeholder algorithm
* an extensive test case infrastructure to test grouping algorithms
* a move of some base resource methods into pgraph refactoring
* some config/compile clean ups to remove code duplication
* b64 encoding/decoding improvements
* a rename of the yaml "res" entries to "kind" (more logical)
* some docs
* small fixes
* and more!
2016-03-28 20:54:41 -04:00
James Shubin
9720812a78 Graph cleanups to make way for the autogroup feature! 2016-03-28 20:54:41 -04:00
James Shubin
05b4066ba6 Add initial plumbing for autogroups
This adds some of the API changes and improvements to the pkg resource
so that it can make use of this feature.
2016-03-28 20:54:41 -04:00
James Shubin
50c458b6cc Support "noarch" style packages in arch check
Forgot about these :)
2016-03-28 20:54:41 -04:00
James Shubin
2ab6d61a61 Avoid errors on golint test with travis
No idea why this happens on travis, and I'm done wasting time trying to
figure it out.
2016-03-28 20:54:41 -04:00
James Shubin
b77a39bdff Add golint test improvements so we detect branches more cleverly
Also fix up a count reversal, hopefully I was just tunnel visioned
before and this is the correct fix!
2016-03-28 05:27:28 -04:00
James Shubin
d3f7432861 Update context since etcd upstream moved to vendor/ dir
Follow suit with what etcd upstream is doing.
2016-03-28 05:27:28 -04:00
Daniele Sluijters
7f3ef5bf85 docs: Add Features and autoedges section 2016-03-14 09:20:53 +01:00
James Shubin
659fb3eb82 Split the Res interface into a Base sub piece
I didn't know this was possible until I was browsing through some golang
docs recently. This should hopefully make it clearer which the common
methods to all resources are (which don't need to be reimplemented each
time) and which ones are unique and need to be created for each
resource.
2016-03-14 01:48:58 -04:00
James Shubin
d1315bb092 Refactor out noop resource into a separate file
I still think it's a useful resource for demonstrating concepts and
perhaps for other future purposes.
2016-03-14 01:28:44 -04:00
James Shubin
b4ac0e2e7c Update README and docs with information about new blog post 2016-03-14 01:03:35 -04:00
James Shubin
bfe619272e Update make deps script to make it better for debian folks
Hopefully this should make it easier for debian users, or for users who
run the script in the wrong directory.
2016-03-11 18:45:28 -05:00
Xavi S.B
963f025011 Fix broken link to graph.yaml section 2016-03-10 20:42:47 +01:00
James Shubin
b8cdcaeb75 Switch bc for awk
This is useful in dumb environments *cough* travis, that apparently
don't have bc as a default :(
2016-03-10 04:23:29 -05:00
James Shubin
6b6dc75152 Add a threshold based golint test case
This lets some golint errors in, but fails if you're over a certain
threshold. The current threshold of 15% (of LOC) is arbitrary and
subject to change. The algorithm should be extended to check a range of
commits, although it's unclear how to detect what range of commits make
up a patch set.
2016-03-10 04:06:57 -05:00
James Shubin
23647445d7 Golint fixes 2016-03-10 03:29:51 -05:00
James Shubin
e60dda5027 Add some of the pkg and svc autoedge logic
This adds another chunk of it, and makes some other small fixes.
2016-03-10 03:29:51 -05:00
James Shubin
f39551952f Add pkg auto edge basics with packagekit improvements
This is a monster patch that finally gets the iterative pkg auto edges
working the way they should. For each file, as soon as one matches, we
don't want to keep add dependencies on other file objects under that
tree structure. This reduces the number of necessary edges considerably,
and allows the graph to run more concurrently.
2016-03-10 03:29:51 -05:00
James Shubin
a9538052bf Split out some of the pkg CheckApply logic
This avoids code duplication since we need to reuse this logic
elsewhere, and as well this adds support for resolving the mapping for
multiple numbers of packages at the same time.

Note that on occasion, InstallPackages has hung waiting for Finished or
Destroy signals which never came! Increasing the PkBufferSize seemed to
help, but it's not clear if this is a permanent fix, or if more advanced
debugging and/or work arounds are needed. Is it possible that not
*waiting* to receive the destroy signals is causing this?
2016-03-10 03:29:51 -05:00
James Shubin
267d5179f5 Exec variable should use pre-existing error variable
Small fixup
2016-03-10 03:29:51 -05:00
James Shubin
10b8c93da4 Make resource "kind" determination more obvious
By adding the "kind" to the base resource, it is still identifiable even
when the resource specific elements are gone in calls that are only
defined in the base resource. This is also more logical for building
resources!

This also switches resources to use an Init() method. This will be
useful for when resources have more complex initialization to do.
2016-03-10 03:29:50 -05:00
James Shubin
c999f0c2cd Add initial "autoedge" plumbing
This allows for resources to automatically add necessary edges to the
graph so that the event system doesn't have to work overtime due to
sub-optimal execution order.
2016-03-10 03:29:50 -05:00
James Shubin
54615dc03b Actually update to 1.6 in travis 2016-03-10 03:22:36 -05:00
James Shubin
9aea95ce85 Bump go versions in travis
Failures might be caused by inotify not working correctly in travis.
This is probably because travis has things mounted over NFS.
https://github.com/travis-ci/travis-ci/issues/2342
2016-02-28 18:44:19 -05:00
James Shubin
80f48291f3 Update Makefile plumbing and associated misc things 2016-02-28 16:57:18 -05:00
James Shubin
1a164cee3e Fix file resource regression
I added a regression to the file resource. This was caused by two
different bugs that I added when I switched the API to use checkapply. I
would have caught these issues, except my test cases *also* had a bug! I
think I've fixed all three issues now.

Lastly, when running on travis, the tests behave very differently! Some
of the tests actually fail, and it's not clear why. As a result, we had
to disable them! I guess you get what you pay for.
2016-02-28 02:18:46 -05:00
James Shubin
da494cdc7c Clean up the examples/ directory
Naming things numerically isn't very obvious. This is better for now.
2016-02-26 02:32:13 -05:00
James Shubin
06635dfa75 Rename GetRes() which is not descriptive, to Kind()
This used to be GetType(), but since now things are "resources", we want
to know what "kind" they are, since asking what "type" they are is
confusing, and makes less logical sense than "Kind".
2016-02-23 00:24:43 -05:00
James Shubin
a56fb3c8cd Update items for Jenkins tests 2016-02-22 20:00:15 -05:00
James Shubin
2dc3c62bbd The svc resource should use the private system bus of course
I didn't realize there was a difference until I was debugging an issue
with the pkg resource, which also uses dbus!
2016-02-22 19:44:10 -05:00
James Shubin
0339d0caa8 Improve go vet testing 2016-02-22 19:43:55 -05:00
James Shubin
3b5678dd91 Add package (pkg) resource
This is based on PackageKit, which means events, *and* we automatically
get support for any of the backends that PackageKit supports. This means
dpkg, and rpm are both first class citizens! Many other backends will
surely work, although thorough testing is left as an exercise to the
reader, or to someone who would like to write more test cases!

Unfortunately at the moment, there are a few upstream PackageKit bugs
which cause us issues, but those have been apparently resolved upstream.
If you experience issues with an old version of PackageKit, test if it
is working correctly before blaming mgmt :)

In parallel, mgmt might increase the testing surface for PackageKit, so
hopefully this makes it more robust for everyone involved!

Lastly, I'd like to point out that many great things that are typically
used for servers do start in the GNOME desktop world. Help support your
GNOME GNU/Linux desktop today!
2016-02-22 19:05:24 -05:00
James Shubin
82ff34234d Fix Makefile warnings on Travis 2016-02-22 01:36:57 -05:00
James Shubin
f3d1369764 Add some fixes for building in tip or with golang 1.6 2016-02-22 01:30:12 -05:00
James Shubin
ed61444d82 Fix golang 1.6 vet issue 2016-02-22 01:05:27 -05:00
James Shubin
ce0d68a8ba Make sure to error on test
This works around set -e not working in this scenario.
2016-02-22 00:59:36 -05:00
James Shubin
74aadbadb8 Update README with small fixes 2016-02-22 00:37:13 -05:00
James Shubin
58f41eddd9 Change API from StateOK/Apply to CheckApply()
This simplifies the API, and reduces code duplication for most
resources. It became obvious when writing the pkg resource, that the two
operations should really be one. Hopefully this will last! Comments
welcome.
2016-02-21 22:07:49 -05:00
James Shubin
4726445ec4 Add great article by felix frank
This is very important work that is doubly hard because the API isn't
stable yet. If you see felix, buy him a beverage.

PS: Sorry felix that I just broke the api. I'll send you the patch to
fix it!
2016-02-21 19:42:21 -05:00
James Shubin
3a85384377 Rename type to resource (res) and service to svc
Naming the resources "type" was a stupid mistake, and is a huge source
of confusion when also talking about real types. Fix this before it gets
out of hand.
2016-02-21 15:51:52 -05:00
James Shubin
d20b529508 Update README with new video
Also did some small formatting fixes
2016-02-20 14:35:52 -05:00
Rob Wilson
7199f558e8 build requires (at least) stringer - add package which contains missing dependency 2016-02-20 10:42:26 +00:00
James Shubin
674cb24a1a Update link to video 2016-02-18 16:00:57 -05:00
James Shubin
02c7336315 Add link to slides from Walter Heck 2016-02-17 01:57:17 -05:00
James Shubin
cde052d819 Add DevConf.cz video recording 2016-02-16 14:26:11 -05:00
James Shubin
989cc8d236 Add more mgmtlove tags 2016-02-16 13:06:22 -05:00
James Shubin
1186d63653 Add a link to my article about debugging golang 2016-02-15 17:37:06 -05:00
James Shubin
6e68d6dda0 Update web articles 2016-02-15 14:18:59 -05:00
James Shubin
7d876701b3 Add test case to match headers 2016-02-14 14:24:26 -05:00
James Shubin
dbbb483853 Switch to CentOS-7.1 for OMV tests.
There is an occasional hang when booting up F23 on Vagrant which causes
jenkins to wait indefinitely. This might be fixed by getting F23 back to
using eth0/eth1 in the base image.
2016-02-12 17:36:51 -05:00
James Shubin
85e9473d56 Make debugging easier when running on Jenkins 2016-02-12 15:55:31 -05:00
James Shubin
427d424707 Update docs to refer to #mgmtlove patches 2016-02-12 15:01:15 -05:00
James Shubin
f90c5fafa4 This now works. w00t 2016-02-12 14:48:00 -05:00
1498 changed files with 183186 additions and 4605 deletions

4
.ackrc
View File

@@ -1,2 +1,6 @@
--ignore-dir=old/
--ignore-dir=tmp/
--ignore-dir=vendor/
--ignore-dir=releases/
--ignore-dir=rpmbuild/
--type-set=mcl:ext:mcl

1
.dockerignore Normal file
View File

@@ -0,0 +1 @@
docker

28
.editorconfig Normal file
View File

@@ -0,0 +1,28 @@
; This file is for unifying the coding style for different editors and IDEs.
; Plugins are available for notepad++, emacs, vim, gedit,
; textmate, visual studio, and more.
;
; See http://editorconfig.org for details.
# Top-most EditorConfig file.
root = true
[*]
end_of_line = lf
insert_final_newline = true
trim_trailing_whitespace = true
[*.sh]
indent_style = tab
[*.go]
indent_style = tab
[Makefile]
indent_style = tab
[*.mcl]
indent_style = tab
[*.txtar]
indent_style = tab

5
.github/FUNDING.yml vendored Normal file
View File

@@ -0,0 +1,5 @@
# You can add one username per supported platform and one custom link.
custom: "https://paypal.me/purpleidea"
github: purpleidea
liberapay: purpleidea
patreon: purpleidea

9
.github/ISSUE_TEMPLATE.md vendored Normal file
View File

@@ -0,0 +1,9 @@
## Versions:
* mgmt version (eg: `mgmt --version`):
* operating system/distribution (eg: `uname -a`):
* golang version (eg: `go version`):
## Description:

47
.github/PULL_REQUEST_TEMPLATE.md vendored Normal file
View File

@@ -0,0 +1,47 @@
## Tips:
* please read the style guide before submitting your patch:
[docs/style-guide.md](../docs/style-guide.md)
* commit message titles must be in the form:
```topic: Capitalized message with no trailing period```
or:
```topic, topic2: Capitalized message with no trailing period```
* golang code must be formatted according to the standard, please run:
```
make gofmt # formats the entire project correctly
```
or format a single golang file correctly:
```
gofmt -w yourcode.go
```
* please rebase your patch against current git master:
```
git checkout master
git pull origin master
git checkout your-feature
git rebase master
git push your-remote your-feature
hub pull-request # or submit with the github web ui
```
* after a patch review, please ping @purpleidea so we know to re-review:
```
# make changes based on reviews...
git add -p # add new changes
git commit --amend # combine with existing commit
git push your-remote your-feature -f
# now ping @purpleidea in the github PR since it doesn't notify us automatically
```
## Thanks for contributing to mgmt and welcome to the team!

98
.github/settings.yml vendored Normal file
View File

@@ -0,0 +1,98 @@
# These settings are synced to GitHub by https://probot.github.io/apps/settings/
repository:
# See https://developer.github.com/v3/repos/#edit for all available settings.
# The name of the repository. Changing this will rename the repository
name: mgmt
# A short description of the repository that will show up on GitHub
description: Next generation distributed, event-driven, parallel config management!
# A URL with more information about the repository
homepage: https://purpleidea.com/tags/mgmtconfig/
# A comma-separated list of topics to set on the repository
topics: golang, go, configuration-management, config-management, devops, etcd, distributed-systems, graph-theory, choreography
# Either `true` to make the repository private, or `false` to make it public.
private: false
# Either `true` to enable issues for this repository, `false` to disable them.
has_issues: true
# Either `true` to enable projects for this repository, or `false` to disable them.
# If projects are disabled for the organization, passing `true` will cause an API error.
has_projects: false
# Either `true` to enable the wiki for this repository, `false` to disable it.
has_wiki: false
# Either `true` to enable downloads for this repository, `false` to disable them.
has_downloads: true
# Updates the default branch for this repository.
default_branch: master
# Either `true` to allow squash-merging pull requests, or `false` to prevent
# squash-merging.
allow_squash_merge: false
# Either `true` to allow merging pull requests with a merge commit, or `false`
# to prevent merging pull requests with merge commits.
allow_merge_commit: false
# Either `true` to allow rebase-merging pull requests, or `false` to prevent
# rebase-merging.
allow_rebase_merge: true
# Labels: define labels for Issues and Pull Requests (in alphabetical order)
labels:
- name: bug
color: fc2929
- name: confirmed
color: d93f0b
- name: design
color: 5319e7
- name: duplicate
color: cccccc
- name: enhancement
color: 84b6eb
- name: good first issue
color: 7057ff
- name: help wanted
color: 159818
- name: invalid
color: e6e6e6
- name: mgmtlove
color: e11d21
- name: question
color: cc317c
- name: needinfo
color: fbca04
- name: wontfix
color: ffffff
# - name: first-timers-only
# # include the old name to rename an existing label
# oldname: Help Wanted
# Collaborators: give specific users access to this repository.
#collaborators:
# - username: purpleidea
# # Note: Only valid on organization-owned repositories.
# # The permission to grant the collaborator. Can be one of:
# # * `pull` - can pull, but not push to or administer this repository.
# # * `push` - can pull and push, but not administer this repository.
# # * `admin` - can pull, push and administer this repository.
# permission: push
# - username: hubot
# permission: pull
# NOTE: The APIs needed for teams are not supported yet by GitHub Apps
# https://developer.github.com/v3/apps/available-endpoints/
#teams:
# - name: core
# permission: admin
# - name: docs
# permission: push

68
.github/workflows/test.yaml vendored Normal file
View File

@@ -0,0 +1,68 @@
# Docs: https://help.github.com/en/articles/workflow-syntax-for-github-actions
# If the name is omitted, it uses the filename instead.
#name: Test
on:
# Run on all pull requests.
pull_request:
#branches:
#- master
# Run on all pushes.
push:
# Run daily at 4am.
schedule:
- cron: 0 4 * * *
jobs:
maketest:
name: Test (${{ matrix.test_block }}) on ${{ matrix.os }} with golang ${{ matrix.golang_version }}
runs-on: ${{ matrix.os }}
env:
GOPATH: /home/runner/work/mgmt/mgmt/go
strategy:
matrix:
# TODO: Add tip when it's supported: https://github.com/actions/setup-go/issues/21
os:
- ubuntu-latest
# macos tests are currently failing in CI
#- macos-latest
golang_version:
# TODO: add 1.24.x and tip
# minimum required and latest published go_version
- "1.23"
test_block:
- basic
- shell
- race
fail-fast: false
steps:
# Do not shallow fetch. The path can't be absolute, so we need to move it
# to the expected location later.
- name: Clone mgmt
uses: actions/checkout@v4
with:
submodules: recursive
fetch-depth: 0
path: ./go/src/github.com/purpleidea/mgmt
- name: Install Go ${{ matrix.golang_version }}
uses: actions/setup-go@v4
with:
go-version: ${{ matrix.golang_version }}
# Install & configure ruby, fixes gem permissions error
- name: Install Ruby
uses: ruby/setup-ruby@v1
with:
ruby-version: head
- name: Install dependencies
working-directory: ./go/src/github.com/purpleidea/mgmt
run: |
make deps
- name: Run test
working-directory: ./go/src/github.com/purpleidea/mgmt
run: |
TEST_BLOCK="${{ matrix.test_block }}" make test

23
.gitignore vendored
View File

@@ -1,9 +1,26 @@
.idea/
.omv/
.ssh/
.vagrant/
mgmt-documentation.pdf
.envrc
old/
tmp/
/vendor/
*WIP
*_stringer.go
mgmt
rpmbuild/
/mgmt
/mgmt.static
# crossbuild artifacts
/build/mgmt-*
mgmt.iml
/rpmbuild/
/releases/
/repository/
/pprof/
/sites/
# vim swap files
.*.sw[op]
# prevent `echo foo 2>1` typo errors by making this file read-only
1
# allow users to keep some junk files around
*.wip

3
.lycheeignore Normal file
View File

@@ -0,0 +1,3 @@
# list URLs that should be excluded for lychee link checher
https://roidelapluie.be
https://github.com/purpleidea/mgmt/commit

View File

@@ -1,25 +0,0 @@
language: go
go:
- 1.4.3
- 1.5.2
- tip
dist: trusty
sudo: required
install: 'make deps'
script: 'make test'
matrix:
allow_failures:
- go: tip
notifications:
irc:
channels:
- "irc.freenode.net#mgmtconfig"
template:
- "%{repository} (%{commit}: %{author}): %{message}"
- "More info : %{build_url}"
on_success: always
on_failure: always
use_notice: false
skip_join: false
email:
- travis-ci@shubin.ca

View File

@@ -1,7 +1,14 @@
This is a list of authors/contributors to the mgmt project.
If you're a contributor, please send a patch with your name.
If you're a core contributor, we might ask you to send a patch with your name.
If you appreciate the work of one of the contributors, thank them a beverage!
For a more exhaustive list please run: git log --format='%aN' | sort -u
This list is sorted alphabetically by first name.
Felix Frank
James Shubin
Joe Groocock
Johan Bloemberg
Jonathan Gold
Julien Pivotto
Paul Morgan
Samuel Gélineau

147
COPYING
View File

@@ -1,21 +1,23 @@
GNU AFFERO GENERAL PUBLIC LICENSE
Version 3, 19 November 2007
GNU GENERAL PUBLIC LICENSE
Version 3, 29 June 2007
Copyright (C) 2007 Free Software Foundation, Inc. <http://fsf.org/>
Copyright (C) 2007 Free Software Foundation, Inc. <https://fsf.org/>
Everyone is permitted to copy and distribute verbatim copies
of this license document, but changing it is not allowed.
Preamble
The GNU Affero General Public License is a free, copyleft license for
software and other kinds of works, specifically designed to ensure
cooperation with the community in the case of network server software.
The GNU General Public License is a free, copyleft license for
software and other kinds of works.
The licenses for most software and other practical works are designed
to take away your freedom to share and change the works. By contrast,
our General Public Licenses are intended to guarantee your freedom to
the GNU General Public License is intended to guarantee your freedom to
share and change all versions of a program--to make sure it remains free
software for all its users.
software for all its users. We, the Free Software Foundation, use the
GNU General Public License for most of our software; it applies also to
any other work released this way by its authors. You can apply it to
your programs, too.
When we speak of free software, we are referring to freedom, not
price. Our General Public Licenses are designed to make sure that you
@@ -24,34 +26,44 @@ them if you wish), that you receive source code or can get it if you
want it, that you can change the software or use pieces of it in new
free programs, and that you know you can do these things.
Developers that use our General Public Licenses protect your rights
with two steps: (1) assert copyright on the software, and (2) offer
you this License which gives you legal permission to copy, distribute
and/or modify the software.
To protect your rights, we need to prevent others from denying you
these rights or asking you to surrender the rights. Therefore, you have
certain responsibilities if you distribute copies of the software, or if
you modify it: responsibilities to respect the freedom of others.
A secondary benefit of defending all users' freedom is that
improvements made in alternate versions of the program, if they
receive widespread use, become available for other developers to
incorporate. Many developers of free software are heartened and
encouraged by the resulting cooperation. However, in the case of
software used on network servers, this result may fail to come about.
The GNU General Public License permits making a modified version and
letting the public access it on a server without ever releasing its
source code to the public.
For example, if you distribute copies of such a program, whether
gratis or for a fee, you must pass on to the recipients the same
freedoms that you received. You must make sure that they, too, receive
or can get the source code. And you must show them these terms so they
know their rights.
The GNU Affero General Public License is designed specifically to
ensure that, in such cases, the modified source code becomes available
to the community. It requires the operator of a network server to
provide the source code of the modified version running there to the
users of that server. Therefore, public use of a modified version, on
a publicly accessible server, gives the public access to the source
code of the modified version.
Developers that use the GNU GPL protect your rights with two steps:
(1) assert copyright on the software, and (2) offer you this License
giving you legal permission to copy, distribute and/or modify it.
An older license, called the Affero General Public License and
published by Affero, was designed to accomplish similar goals. This is
a different license, not a version of the Affero GPL, but Affero has
released a new version of the Affero GPL which permits relicensing under
this license.
For the developers' and authors' protection, the GPL clearly explains
that there is no warranty for this free software. For both users' and
authors' sake, the GPL requires that modified versions be marked as
changed, so that their problems will not be attributed erroneously to
authors of previous versions.
Some devices are designed to deny users access to install or run
modified versions of the software inside them, although the manufacturer
can do so. This is fundamentally incompatible with the aim of
protecting users' freedom to change the software. The systematic
pattern of such abuse occurs in the area of products for individuals to
use, which is precisely where it is most unacceptable. Therefore, we
have designed this version of the GPL to prohibit the practice for those
products. If such problems arise substantially in other domains, we
stand ready to extend this provision to those domains in future versions
of the GPL, as needed to protect the freedom of users.
Finally, every program is threatened constantly by software patents.
States should not allow patents to restrict development and use of
software on general-purpose computers, but in those that do, we wish to
avoid the special danger that patents applied to a free program could
make it effectively proprietary. To prevent this, the GPL assures that
patents cannot be used to render the program non-free.
The precise terms and conditions for copying, distribution and
modification follow.
@@ -60,7 +72,7 @@ modification follow.
0. Definitions.
"This License" refers to version 3 of the GNU Affero General Public License.
"This License" refers to version 3 of the GNU General Public License.
"Copyright" also means copyright-like laws that apply to other kinds of
works, such as semiconductor masks.
@@ -537,45 +549,35 @@ to collect a royalty for further conveying from those to whom you convey
the Program, the only way you could satisfy both those terms and this
License would be to refrain entirely from conveying the Program.
13. Remote Network Interaction; Use with the GNU General Public License.
Notwithstanding any other provision of this License, if you modify the
Program, your modified version must prominently offer all users
interacting with it remotely through a computer network (if your version
supports such interaction) an opportunity to receive the Corresponding
Source of your version by providing access to the Corresponding Source
from a network server at no charge, through some standard or customary
means of facilitating copying of software. This Corresponding Source
shall include the Corresponding Source for any work covered by version 3
of the GNU General Public License that is incorporated pursuant to the
following paragraph.
13. Use with the GNU Affero General Public License.
Notwithstanding any other provision of this License, you have
permission to link or combine any covered work with a work licensed
under version 3 of the GNU General Public License into a single
under version 3 of the GNU Affero General Public License into a single
combined work, and to convey the resulting work. The terms of this
License will continue to apply to the part which is the covered work,
but the work with which it is combined will remain governed by version
3 of the GNU General Public License.
but the special requirements of the GNU Affero General Public License,
section 13, concerning interaction through a network will apply to the
combination as such.
14. Revised Versions of this License.
The Free Software Foundation may publish revised and/or new versions of
the GNU Affero General Public License from time to time. Such new versions
will be similar in spirit to the present version, but may differ in detail to
the GNU General Public License from time to time. Such new versions will
be similar in spirit to the present version, but may differ in detail to
address new problems or concerns.
Each version is given a distinguishing version number. If the
Program specifies that a certain numbered version of the GNU Affero General
Program specifies that a certain numbered version of the GNU General
Public License "or any later version" applies to it, you have the
option of following the terms and conditions either of that numbered
version or of any later version published by the Free Software
Foundation. If the Program does not specify a version number of the
GNU Affero General Public License, you may choose any version ever published
GNU General Public License, you may choose any version ever published
by the Free Software Foundation.
If the Program specifies that a proxy can decide which future
versions of the GNU Affero General Public License can be used, that proxy's
versions of the GNU General Public License can be used, that proxy's
public statement of acceptance of a version permanently authorizes you
to choose that version for the Program.
@@ -633,29 +635,40 @@ the "copyright" line and a pointer to where the full notice is found.
Copyright (C) <year> <name of author>
This program is free software: you can redistribute it and/or modify
it under the terms of the GNU Affero General Public License as published by
it under the terms of the GNU General Public License as published by
the Free Software Foundation, either version 3 of the License, or
(at your option) any later version.
This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU Affero General Public License for more details.
GNU General Public License for more details.
You should have received a copy of the GNU Affero General Public License
along with this program. If not, see <http://www.gnu.org/licenses/>.
You should have received a copy of the GNU General Public License
along with this program. If not, see <https://www.gnu.org/licenses/>.
Also add information on how to contact you by electronic and paper mail.
If your software can interact with users remotely through a computer
network, you should also make sure that it provides a way for users to
get its source. For example, if your program is a web application, its
interface could display a "Source" link that leads users to an archive
of the code. There are many ways you could offer source, and different
solutions will be better for different programs; see section 13 for the
specific requirements.
If the program does terminal interaction, make it output a short
notice like this when it starts in an interactive mode:
<program> Copyright (C) <year> <name of author>
This program comes with ABSOLUTELY NO WARRANTY; for details type `show w'.
This is free software, and you are welcome to redistribute it
under certain conditions; type `show c' for details.
The hypothetical commands `show w' and `show c' should show the appropriate
parts of the General Public License. Of course, your program's commands
might be different; for a GUI interface, you would use an "about box".
You should also get your employer (if you work as a programmer) or school,
if any, to sign a "copyright disclaimer" for the program, if necessary.
For more information on this, and how to apply and follow the GNU AGPL, see
<http://www.gnu.org/licenses/>.
For more information on this, and how to apply and follow the GNU GPL, see
<https://www.gnu.org/licenses/>.
The GNU General Public License does not permit incorporating your program
into proprietary programs. If your program is a subroutine library, you
may consider it more useful to permit linking proprietary applications with
the library. If this is what you want to do, use the GNU Lesser General
Public License instead of this License. But first, please read
<https://www.gnu.org/philosophy/why-not-lgpl.html>.

View File

@@ -1,16 +1,28 @@
Mgmt
Copyright (C) 2013-2016+ James Shubin and the project contributors
Copyright (C) James Shubin and the project contributors
Written by James Shubin <james@shubin.ca> and the project contributors
This program is free software: you can redistribute it and/or modify
it under the terms of the GNU Affero General Public License as published by
it under the terms of the GNU General Public License as published by
the Free Software Foundation, either version 3 of the License, or
(at your option) any later version.
This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU Affero General Public License for more details.
GNU General Public License for more details.
You should have received a copy of the GNU Affero General Public License
along with this program. If not, see <http://www.gnu.org/licenses/>.
You should have received a copy of the GNU General Public License
along with this program. If not, see <https://www.gnu.org/licenses/>.
Additional permission under GNU GPL version 3 section 7
If you modify this program, or any covered work, by linking or combining it
with embedded mcl code and modules (and that the embedded mcl code and
modules which link with this program, contain a copy of their source code in
the authoritative form) containing parts covered by the terms of any other
license, the licensors of this program grant you additional permission to
convey the resulting work. Furthermore, the licensors of this program grant
the original author, James Shubin, additional permission to update this
additional permission if he deems it necessary to achieve the goals of this
additional permission.

View File

@@ -1,143 +0,0 @@
#mgmt
<!--
Mgmt
Copyright (C) 2013-2016+ James Shubin and the project contributors
Written by James Shubin <james@shubin.ca> and the project contributors
This program is free software: you can redistribute it and/or modify
it under the terms of the GNU Affero General Public License as published by
the Free Software Foundation, either version 3 of the License, or
(at your option) any later version.
This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU Affero General Public License for more details.
You should have received a copy of the GNU Affero General Public License
along with this program. If not, see <http://www.gnu.org/licenses/>.
-->
##mgmt by [James](https://ttboj.wordpress.com/)
####Available from:
####[https://github.com/purpleidea/mgmt/](https://github.com/purpleidea/mgmt/)
####This documentation is available in: [Markdown](https://github.com/purpleidea/mgmt/blob/master/DOCUMENTATION.md) or [PDF](https://pdfdoc-purpleidea.rhcloud.com/pdf/https://github.com/purpleidea/mgmt/blob/master/DOCUMENTATION.md) format.
####Table of Contents
1. [Overview](#overview)
2. [Project description - What the project does](#project-description)
3. [Setup - Getting started with mgmt](#setup)
4. [Usage/FAQ - Notes on usage and frequently asked questions](#usage-and-frequently-asked-questions)
5. [Reference - Detailed reference](#reference)
* [graph.yaml](#graph.yaml)
* [Command line](#command-line)
6. [Examples - Example configurations](#examples)
7. [Development - Background on module development and reporting bugs](#development)
8. [Authors - Authors and contact information](#authors)
##Overview
The `mgmt` tool is a research prototype to demonstrate next generation config
management techniques. Hopefully it will evolve into a useful, robust tool.
##Project Description
The mgmt tool is a distributed, event driven, config management tool, that
supports parallel execution, and librarification to be used as the management
foundation in and for, new and existing software.
##Setup
During this prototype phase, the tool can be run out of the source directory.
You'll probably want to use ```./run.sh run --file examples/graph1.yaml``` to
get started. Beware that this _can_ cause data loss. Understand what you're
doing first, or perform these actions in a virtual environment such as the one
provided by [Oh-My-Vagrant](https://github.com/purpleidea/oh-my-vagrant).
##Usage and frequently asked questions
(Send your questions as a patch to this FAQ! I'll review it, merge it, and
respond by commit with the answer.)
###Why did you start this project?
I wanted a next generation config management solution that didn't have all of
the design flaws or limitations that the current generation of tools do, and no
tool existed!
###Why did you use etcd? What about consul?
Etcd and consul are both written in golang, which made them the top two
contenders for my prototype. Ultimately a choice had to be made, and etcd was
chosen, but it was also somewhat arbitrary. If there is available interest,
good reasoning, *and* patches, then we would consider either switching or
supporting both, but this is not a high priority at this time.
###You didn't answer my question, or I have a question!
It's best to ask on [IRC](https://webchat.freenode.net/?channels=#mgmtconfig)
to see if someone can help you. Once we get a big enough community going, we'll
add a mailing list. If you don't get any response from the above, you can
contact me through my [technical blog](https://ttboj.wordpress.com/contact/)
and I'll do my best to help. If you have a good question, please add it as a
patch to this documentation. I'll merge your question, and add a patch with the
answer!
##Reference
Please note that there are a number of undocumented options. For more
information on these options, please view the source at:
[https://github.com/purpleidea/mgmt/](https://github.com/purpleidea/mgmt/).
If you feel that a well used option needs documenting here, please patch it!
###Overview of reference
* [graph.yaml](#graph.yaml): Main graph definition file.
* [Command line](#command-line): Command line parameters.
###graph.yaml
This is the compiled graph definition file. The format is currently
undocumented, but by looking through the [examples/](https://github.com/purpleidea/mgmt/tree/master/examples)
you can probably figure out most of it, as it's fairly intuitive.
###Command line
The main interface to the `mgmt` tool is the command line. For the most recent
documentation, please run `mgmt --help`.
####`--file <graph.yaml>`
Point to a graph file to run.
####`--converged-timeout <seconds>`
Exit if the machine has converged for approximately this many seconds.
####`--max-runtime <seconds>`
Exit when the agent has run for approximately this many seconds. This is not
generally recommended, but may be useful for users who know what they're doing.
##Examples
For example configurations, please consult the [examples/](https://github.com/purpleidea/mgmt/tree/master/examples) directory in the git
source repository. It is available from:
[https://github.com/purpleidea/mgmt/tree/master/examples](https://github.com/purpleidea/mgmt/tree/master/examples)
##Development
This is a project that I started in my free time in 2013. Development is driven
by all of our collective patches! Dive right in, and start hacking!
Please contact me if you'd like to invite me to speak about this at your event.
You can follow along [on my technical blog](https://ttboj.wordpress.com/).
To report any bugs, please file a ticket at: [https://github.com/purpleidea/mgmt/issues](https://github.com/purpleidea/mgmt/issues).
##Authors
Copyright (C) 2013-2016+ James Shubin and the project contributors
Please see the
[AUTHORS](https://github.com/purpleidea/mgmt/tree/master/AUTHORS) file
for more information.
* [github](https://github.com/purpleidea/)
* [&#64;purpleidea](https://twitter.com/#!/purpleidea)
* [https://ttboj.wordpress.com/](https://ttboj.wordpress.com/)

549
Makefile
View File

@@ -1,91 +1,290 @@
# Mgmt
# Copyright (C) 2013-2016+ James Shubin and the project contributors
# Copyright (C) James Shubin and the project contributors
# Written by James Shubin <james@shubin.ca> and the project contributors
#
# This program is free software: you can redistribute it and/or modify
# it under the terms of the GNU Affero General Public License as published by
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU Affero General Public License for more details.
# GNU General Public License for more details.
#
# You should have received a copy of the GNU Affero General Public License
# along with this program. If not, see <http://www.gnu.org/licenses/>.
# You should have received a copy of the GNU General Public License
# along with this program. If not, see <https://www.gnu.org/licenses/>.
#
# Additional permission under GNU GPL version 3 section 7
#
# If you modify this program, or any covered work, by linking or combining it
# with embedded mcl code and modules (and that the embedded mcl code and
# modules which link with this program, contain a copy of their source code in
# the authoritative form) containing parts covered by the terms of any other
# license, the licensors of this program grant you additional permission to
# convey the resulting work. Furthermore, the licensors of this program grant
# the original author, James Shubin, additional permission to update this
# additional permission if he deems it necessary to achieve the goals of this
# additional permission.
SHELL = /bin/bash
.PHONY: all version program path deps run race build clean test format docs rpmbuild rpm srpm spec tar upload upload-sources upload-srpms upload-rpms copr
SHELL = bash
.PHONY: all art cleanart version program lang path deps run race generate build build-debug crossbuild clean test gofmt yamlfmt format docs
.PHONY: rpmbuild mkdirs rpm srpm spec tar upload upload-sources upload-srpms upload-rpms upload-releases copr tag
.PHONY: mkosi mkosi_fedora-latest mkosi_fedora-older mkosi_stream-latest mkosi_debian-stable mkosi_ubuntu-latest mkosi_archlinux
.PHONY: release release_test releases_path release_binary_amd64 release_binary_arm64 release_fedora-latest release_fedora-older release_stream-latest release_debian-stable release_ubuntu-latest release_archlinux
.PHONY: funcgen
.SILENT: clean
SVERSION := $(shell git describe --match '[0-9]*\.[0-9]*\.[0-9]*' --tags --dirty --always)
VERSION := $(shell git describe --match '[0-9]*\.[0-9]*\.[0-9]*' --tags --abbrev=0)
PROGRAM := $(shell basename --suffix=-$(VERSION) $(notdir $(CURDIR)))
# a large amount of output from this `find`, can cause `make` to be much slower!
GO_FILES := $(shell find * -name '*.go' -not -path 'old/*' -not -path 'tmp/*')
MCL_FILES := $(shell find lang/ -name '*.mcl' -not -path 'old/*' -not -path 'tmp/*')
MISC_FILES := $(shell find engine/resources/http_server_ui/)
SVERSION := $(or $(SVERSION),$(shell git describe --match '[0-9]*\.[0-9]*\.[0-9]*' --tags --dirty --always))
VERSION := $(or $(VERSION),$(shell git describe --match '[0-9]*\.[0-9]*\.[0-9]*' --tags --abbrev=0))
PROGRAM := $(shell echo $(notdir $(CURDIR)) | cut -f1 -d"-")
PKGNAME := $(shell go list .)
ifeq ($(VERSION),$(SVERSION))
RELEASE = 1
else
RELEASE = untagged
endif
ARCH = $(shell arch)
SPEC = rpmbuild/SPECS/mgmt.spec
SOURCE = rpmbuild/SOURCES/mgmt-$(VERSION).tar.bz2
SRPM = rpmbuild/SRPMS/mgmt-$(VERSION)-$(RELEASE).src.rpm
SRPM_BASE = mgmt-$(VERSION)-$(RELEASE).src.rpm
RPM = rpmbuild/RPMS/mgmt-$(VERSION)-$(RELEASE).$(ARCH).rpm
# debugging is harder if -trimpath is set, so disable it if env var is set
# this is force enabled automatically when using the release target
ifeq ($(MGMT_NOTRIMPATH),true)
TRIMPATH =
else
TRIMPATH = -trimpath
endif
ARCH = $(uname -m)
SPEC = rpmbuild/SPECS/$(PROGRAM).spec
SOURCE = rpmbuild/SOURCES/$(PROGRAM)-$(VERSION).tar.bz2
SRPM = rpmbuild/SRPMS/$(PROGRAM)-$(VERSION)-$(RELEASE).src.rpm
SRPM_BASE = $(PROGRAM)-$(VERSION)-$(RELEASE).src.rpm
RPM = rpmbuild/RPMS/$(PROGRAM)-$(VERSION)-$(RELEASE).$(ARCH).rpm
USERNAME := $(shell cat ~/.config/copr 2>/dev/null | grep username | awk -F '=' '{print $$2}' | tr -d ' ')
SERVER = 'dl.fedoraproject.org'
REMOTE_PATH = 'pub/alt/$(USERNAME)/mgmt'
REMOTE_PATH = '/srv/pub/alt/$(USERNAME)/$(PROGRAM)'
ifneq ($(GOTAGS),)
BUILD_FLAGS = -tags '$(GOTAGS)'
endif
GOOSARCHES ?= linux/amd64 linux/ppc64 linux/ppc64le linux/arm64 darwin/amd64
all: docs
GOHOSTOS = $(shell go env GOHOSTOS)
GOHOSTARCH = $(shell go env GOHOSTARCH)
# The underscores separate the prefix name ("TOKEN") the distro ("BINARY",
# "FEDORA-LATEST", etc...) and the arch ("AMD64"). The distro can have a dash.
TOKEN_BINARY_AMD64 = $(shell grep -v '#' releases/binary-amd64.release)
TOKEN_BINARY_ARM64 = $(shell grep -v '#' releases/binary-arm64.release)
TOKEN_FEDORA-LATEST = $(shell grep -v '#' releases/fedora-latest.release)
TOKEN_FEDORA-OLDER = $(shell grep -v '#' releases/fedora-older.release)
TOKEN_STREAM-LATEST = $(shell grep -v '#' releases/stream-latest.release)
TOKEN_DEBIAN-STABLE = $(shell grep -v '#' releases/debian-stable.release)
TOKEN_UBUNTU-LATEST = $(shell grep -v '#' releases/ubuntu-latest.release)
TOKEN_ARCHLINUX = $(shell grep -v '#' releases/archlinux.release)
FILE_BINARY_AMD64 = mgmt-linux-amd64-$(VERSION)
FILE_BINARY_ARM64 = mgmt-linux-arm64-$(VERSION)
# TODO: add ARCH onto these at the end, eg _AMD64
FILE_FEDORA-LATEST = mgmt-$(TOKEN_FEDORA-LATEST)-$(VERSION)-1.x86_64.rpm
FILE_FEDORA-OLDER = mgmt-$(TOKEN_FEDORA-OLDER)-$(VERSION)-1.x86_64.rpm
FILE_STREAM-LATEST = mgmt-$(TOKEN_STREAM-LATEST)-$(VERSION)-1.x86_64.rpm
FILE_DEBIAN-STABLE = mgmt_$(TOKEN_DEBIAN-STABLE)_$(VERSION)_amd64.deb
FILE_UBUNTU-LATEST = mgmt_$(TOKEN_UBUNTU-LATEST)_$(VERSION)_amd64.deb
FILE_ARCHLINUX = mgmt-$(TOKEN_ARCHLINUX)-$(VERSION)-1-x86_64.pkg.tar.xz
PKG_BINARY_AMD64 = releases/$(VERSION)/$(TOKEN_BINARY_AMD64)/$(FILE_BINARY_AMD64)
PKG_BINARY_ARM64 = releases/$(VERSION)/$(TOKEN_BINARY_ARM64)/$(FILE_BINARY_ARM64)
PKG_FEDORA-LATEST = releases/$(VERSION)/$(TOKEN_FEDORA-LATEST)/$(FILE_FEDORA-LATEST)
PKG_FEDORA-OLDER = releases/$(VERSION)/$(TOKEN_FEDORA-OLDER)/$(FILE_FEDORA-OLDER)
PKG_STREAM-LATEST = releases/$(VERSION)/$(TOKEN_STREAM-LATEST)/$(FILE_STREAM-LATEST)
PKG_DEBIAN-STABLE = releases/$(VERSION)/$(TOKEN_DEBIAN-STABLE)/$(FILE_DEBIAN-STABLE)
PKG_UBUNTU-LATEST = releases/$(VERSION)/$(TOKEN_UBUNTU-LATEST)/$(FILE_UBUNTU-LATEST)
PKG_ARCHLINUX = releases/$(VERSION)/$(TOKEN_ARCHLINUX)/$(FILE_ARCHLINUX)
DEP_BINARY_AMD64 =
ifneq ($(TOKEN_BINARY_AMD64),)
DEP_BINARY_AMD64 = $(PKG_BINARY_AMD64)
endif
DEP_BINARY_ARM64 =
ifneq ($(TOKEN_BINARY_ARM64),)
DEP_BINARY_ARM64 = $(PKG_BINARY_ARM64)
endif
DEP_FEDORA-LATEST =
ifneq ($(TOKEN_FEDORA-LATEST),)
DEP_FEDORA-LATEST = $(PKG_FEDORA-LATEST)
endif
DEP_FEDORA-OLDER =
ifneq ($(TOKEN_FEDORA-OLDER),)
DEP_FEDORA-OLDER = $(PKG_FEDORA-OLDER)
endif
DEP_STREAM-LATEST =
ifneq ($(TOKEN_STREAM-LATEST),)
DEP_STREAM-LATEST = $(PKG_STREAM-LATEST)
endif
DEP_DEBIAN-STABLE =
ifneq ($(TOKEN_DEBIAN-STABLE),)
DEP_DEBIAN-STABLE = $(PKG_DEBIAN-STABLE)
endif
DEP_UBUNTU-LATEST =
ifneq ($(TOKEN_UBUNTU-LATEST),)
DEP_UBUNTU-LATEST = $(PKG_UBUNTU-LATEST)
endif
DEP_ARCHLINUX =
ifneq ($(TOKEN_ARCHLINUX),)
DEP_ARCHLINUX = $(PKG_ARCHLINUX)
endif
SHA256SUMS = releases/$(VERSION)/SHA256SUMS
SHA256SUMS_ASC = $(SHA256SUMS).asc
default: build
#
# art
#
art: art/mgmt_logo_default_symbol.png art/mgmt_logo_default_tall.png art/mgmt_logo_default_wide.png art/mgmt_logo_reversed_symbol.png art/mgmt_logo_reversed_tall.png art/mgmt_logo_reversed_wide.png art/mgmt_logo_white_symbol.png art/mgmt_logo_white_tall.png art/mgmt_logo_white_wide.png ## generate artwork
cleanart:
rm -f art/mgmt_logo_default_symbol.png art/mgmt_logo_default_tall.png art/mgmt_logo_default_wide.png art/mgmt_logo_reversed_symbol.png art/mgmt_logo_reversed_tall.png art/mgmt_logo_reversed_wide.png art/mgmt_logo_white_symbol.png art/mgmt_logo_white_tall.png art/mgmt_logo_white_wide.png
# NOTE: the widths are arbitrary
art/mgmt_logo_default_symbol.png: art/mgmt_logo_default_symbol.svg
inkscape --export-background='#ffffff' --without-gui --export-png "$@" --export-width 300 $(@:png=svg)
art/mgmt_logo_default_tall.png: art/mgmt_logo_default_tall.svg
inkscape --export-background='#ffffff' --without-gui --export-png "$@" --export-width 400 $(@:png=svg)
art/mgmt_logo_default_wide.png: art/mgmt_logo_default_wide.svg
inkscape --export-background='#ffffff' --without-gui --export-png "$@" --export-width 800 $(@:png=svg)
art/mgmt_logo_reversed_symbol.png: art/mgmt_logo_reversed_symbol.svg
inkscape --export-background='#231f20' --without-gui --export-png "$@" --export-width 300 $(@:png=svg)
art/mgmt_logo_reversed_tall.png: art/mgmt_logo_reversed_tall.svg
inkscape --export-background='#231f20' --without-gui --export-png "$@" --export-width 400 $(@:png=svg)
art/mgmt_logo_reversed_wide.png: art/mgmt_logo_reversed_wide.svg
inkscape --export-background='#231f20' --without-gui --export-png "$@" --export-width 800 $(@:png=svg)
art/mgmt_logo_white_symbol.png: art/mgmt_logo_white_symbol.svg
inkscape --export-background='#231f20' --without-gui --export-png "$@" --export-width 300 $(@:png=svg)
art/mgmt_logo_white_tall.png: art/mgmt_logo_white_tall.svg
inkscape --export-background='#231f20' --without-gui --export-png "$@" --export-width 400 $(@:png=svg)
art/mgmt_logo_white_wide.png: art/mgmt_logo_white_wide.svg
inkscape --export-background='#231f20' --without-gui --export-png "$@" --export-width 800 $(@:png=svg)
all: docs $(PROGRAM).static
# show the current version
version:
version: ## show the current version
@echo $(VERSION)
program:
program: ## show the program name
@echo $(PROGRAM)
path:
path: ## create working paths
./misc/make-path.sh
deps:
deps: ## install system and golang dependencies
./misc/make-deps.sh
run:
find -maxdepth 1 -type f -name '*.go' -not -name '*_test.go' | xargs go run -ldflags "-X main.program=$(PROGRAM) -X main.version=$(SVERSION)"
# include race flag
race:
find -maxdepth 1 -type f -name '*.go' -not -name '*_test.go' | xargs go run -race -ldflags "-X main.program=$(PROGRAM) -X main.version=$(SVERSION)"
build: mgmt
mgmt: main.go
@echo "Building: $(PROGRAM), version: $(SVERSION)."
generate:
go generate
# avoid equals sign in old golang versions eg in: -X foo=bar
if go version | grep -qE 'go1.3|go1.4'; then \
go build -ldflags "-X main.program $(PROGRAM) -X main.version $(SVERSION)" -o mgmt; \
else \
go build -ldflags "-X main.program=$(PROGRAM) -X main.version=$(SVERSION)" -o mgmt; \
fi
clean:
[ ! -e mgmt ] || rm mgmt
lang: ## generates the lexer/parser for the language frontend
@# recursively run make in child dir named lang
@$(MAKE) --quiet -C lang
resources: ## builds the resources dependencies required for the engine backend
@# recursively run make in child dir named engine/resources
@$(MAKE) --quiet -C engine/resources
# build a `mgmt` binary for current host os/arch
$(PROGRAM): build/mgmt-${GOHOSTOS}-${GOHOSTARCH} ## build an mgmt binary for current host os/arch
cp -a $< $@
$(PROGRAM).static: $(GO_FILES) $(MCL_FILES) $(MISC_FILES) go.mod go.sum
@echo "Building: $(PROGRAM).static, version: $(SVERSION)..."
go generate
go build $(TRIMPATH) -a -installsuffix cgo -tags netgo -ldflags '-extldflags "-static" -X main.program=$(PROGRAM) -X main.version=$(SVERSION) -s -w' -o $(PROGRAM).static $(BUILD_FLAGS);
build: LDFLAGS=-s -w ## build a fresh mgmt binary
build: $(PROGRAM)
build-debug: LDFLAGS=
build-debug: $(PROGRAM)
# if you're using the bad/dev branch, you might want this too!
baddev: BUILD_FLAGS = -tags 'noaugeas novirt'
baddev: $(PROGRAM)
# pattern rule target for (cross)building, mgmt-OS-ARCH will be expanded to the correct build
# extract os and arch from target pattern
GOOS=$(firstword $(subst -, ,$*))
GOARCH=$(lastword $(subst -, ,$*))
build/mgmt-%: $(GO_FILES) $(MCL_FILES) $(MISC_FILES) go.mod go.sum | lang resources funcgen
@# If you need to run `go mod tidy` then this can trigger.
@if [ "$(PKGNAME)" = "" ]; then echo "\$$(PKGNAME) is empty, test with: go list ."; exit 42; fi
@echo "Building: $(PROGRAM), os/arch: $*, version: $(SVERSION)..."
@# XXX: leave race detector on by default for now. For production
@# builds, we can consider turning it off for performance improvements.
@# XXX: ./mgmt run --tmp-prefix lang something_fast.mcl > /tmp/race 2>&1 # search for "WARNING: DATA RACE"
time env GOOS=${GOOS} GOARCH=${GOARCH} go build $(TRIMPATH) -race -ldflags=$(PKGNAME)="-X main.program=$(PROGRAM) -X main.version=$(SVERSION) ${LDFLAGS}" -o $@ $(BUILD_FLAGS)
# create a list of binary file names to use as make targets
# to use this you might want to run something like:
# GOOSARCHES='linux/arm64' GOTAGS='noaugeas novirt' make crossbuild
# and the output will end up in build/
crossbuild_targets = $(addprefix build/mgmt-,$(subst /,-,${GOOSARCHES}))
crossbuild: ${crossbuild_targets}
clean: ## clean things up
$(MAKE) --quiet -C test clean
$(MAKE) --quiet -C lang clean
$(MAKE) --quiet -C engine/resources clean
$(MAKE) --quiet -C misc/mkosi clean
rm -f lang/core/generated_funcs.go || true
rm -f lang/core/generated_funcs_test.go || true
[ ! -e $(PROGRAM) ] || rm $(PROGRAM)
rm -f *_stringer.go # generated by `go generate`
rm -f *_mock.go # generated by `go generate`
# crossbuild artifacts
rm -f build/mgmt-*
test:
test: build ## run tests
@# recursively run make in child dir named test
@$(MAKE) --quiet -C test
./test.sh
format:
find -type f -name '*.go' -not -path './old/*' -not -path './tmp/*' -exec gofmt -w {} \;
find -type f -name '*.yaml' -not -path './old/*' -not -path './tmp/*' -not -path './omv.yaml' -exec ruby -e "require 'yaml'; x=YAML.load_file('{}').to_yaml.each_line.map(&:rstrip).join(10.chr)+10.chr; File.open('{}', 'w').write x" \;
# create all test targets for make tab completion (eg: make test-gofmt)
test_suites=$(shell find test/ -maxdepth 1 -name test-* -exec basename {} .sh \;)
# allow to run only one test suite at a time
${test_suites}: test-%: build
./test.sh $*
docs: mgmt-documentation.pdf
# targets to run individual shell tests (eg: make test-shell-load0)
test_shell=$(shell find test/shell/ -maxdepth 1 -name "*.sh" -exec basename {} .sh \;)
$(addprefix test-shell-,${test_shell}): test-shell-%: build
./test/test-shell.sh "$*.sh"
mgmt-documentation.pdf: DOCUMENTATION.md
pandoc DOCUMENTATION.md -o 'mgmt-documentation.pdf'
gofmt:
# TODO: remove gofmt once goimports has a -s option
find . -maxdepth 9 -type f -name '*.go' -not -path './old/*' -not -path './tmp/*' -not -path './vendor/*' -exec gofmt -s -w {} \;
find . -maxdepth 9 -type f -name '*.go' -not -path './old/*' -not -path './tmp/*' -not -path './vendor/*' -exec goimports -w {} \;
yamlfmt:
find . -maxdepth 3 -type f -name '*.yaml' -not -path './old/*' -not -path './tmp/*' -not -path './omv.yaml' -exec ruby -e "require 'yaml'; x=YAML.load_file('{}').to_yaml.each_line.map(&:rstrip).join(10.chr)+10.chr; File.open('{}', 'w').write x" \;
format: gofmt yamlfmt ## format yaml and golang code
docs: $(PROGRAM)-documentation.pdf ## generate docs
$(PROGRAM)-documentation.pdf: docs/documentation.md
pandoc docs/documentation.md -o docs/'$(PROGRAM)-documentation.pdf'
#
# build aliases
@@ -107,7 +306,7 @@ rpmbuild/SOURCES/: tar
rpmbuild/SRPMS/: srpm
rpmbuild/RPMS/: rpm
upload: upload-sources upload-srpms upload-rpms
upload: upload-sources upload-srpms upload-rpms ## upload sources
# do nothing
#
@@ -116,21 +315,21 @@ upload: upload-sources upload-srpms upload-rpms
$(RPM): $(SPEC) $(SOURCE)
@echo Running rpmbuild -bb...
rpmbuild --define '_topdir $(shell pwd)/rpmbuild' -bb $(SPEC) && \
mv rpmbuild/RPMS/$(ARCH)/mgmt-$(VERSION)-$(RELEASE).*.rpm $(RPM)
mv rpmbuild/RPMS/$(ARCH)/$(PROGRAM)-$(VERSION)-$(RELEASE).*.rpm $(RPM)
$(SRPM): $(SPEC) $(SOURCE)
@echo Running rpmbuild -bs...
rpmbuild --define '_topdir $(shell pwd)/rpmbuild' -bs $(SPEC)
# renaming is not needed because we aren't using the dist variable
#mv rpmbuild/SRPMS/mgmt-$(VERSION)-$(RELEASE).*.src.rpm $(SRPM)
#mv rpmbuild/SRPMS/$(PROGRAM)-$(VERSION)-$(RELEASE).*.src.rpm $(SRPM)
#
# spec
#
$(SPEC): rpmbuild/ mgmt.spec.in
$(SPEC): rpmbuild/ spec.in
@echo Running templater...
#cat mgmt.spec.in > $(SPEC)
sed -e s/__VERSION__/$(VERSION)/ -e s/__RELEASE__/$(RELEASE)/ < mgmt.spec.in > $(SPEC)
cat spec.in > $(SPEC)
sed -e s/__PROGRAM__/$(PROGRAM)/g -e s/__VERSION__/$(VERSION)/g -e s/__RELEASE__/$(RELEASE)/g < spec.in > $(SPEC)
# append a changelog to the .spec file
git log --format="* %cd %aN <%aE>%n- (%h) %s%d%n" --date=local | sed -r 's/[0-9]+:[0-9]+:[0-9]+ //' >> $(SPEC)
@@ -140,7 +339,7 @@ $(SPEC): rpmbuild/ mgmt.spec.in
$(SOURCE): rpmbuild/
@echo Running git archive...
# use HEAD if tag doesn't exist yet, so that development is easier...
git archive --prefix=mgmt-$(VERSION)/ -o $(SOURCE) $(VERSION) 2> /dev/null || (echo 'Warning: $(VERSION) does not exist. Using HEAD instead.' && git archive --prefix=mgmt-$(VERSION)/ -o $(SOURCE) HEAD)
git archive --prefix=$(PROGRAM)-$(VERSION)/ -o $(SOURCE) $(VERSION) 2> /dev/null || (echo 'Warning: $(VERSION) does not exist. Using HEAD instead.' && git archive --prefix=$(PROGRAM)-$(VERSION)/ -o $(SOURCE) HEAD)
# TODO: if git archive had a --submodules flag this would easier!
@echo Running git archive submodules...
# i thought i would need --ignore-zeros, but it doesn't seem necessary!
@@ -149,14 +348,14 @@ $(SOURCE): rpmbuild/
temp="$${temp#\'}"; \
path=$$temp; \
[ "$$path" = "" ] && continue; \
(cd $$path && git archive --prefix=mgmt-$(VERSION)/$$path/ HEAD > $$p/rpmbuild/tmp.tar && tar --concatenate --file=$$p/$(SOURCE) $$p/rpmbuild/tmp.tar && rm $$p/rpmbuild/tmp.tar); \
(cd $$path && git archive --prefix=$(PROGRAM)-$(VERSION)/$$path/ HEAD > $$p/rpmbuild/tmp.tar && tar --concatenate --file=$$p/$(SOURCE) $$p/rpmbuild/tmp.tar && rm $$p/rpmbuild/tmp.tar); \
done
# TODO: ensure that each sub directory exists
rpmbuild/:
mkdir -p rpmbuild/{BUILD,BUILDROOT,RPMS,SOURCES,SPECS,SRPMS}
rpmbuild:
mkdirs:
mkdir -p rpmbuild/{BUILD,BUILDROOT,RPMS,SOURCES,SPECS,SRPMS}
#
@@ -212,10 +411,244 @@ upload-rpms: rpmbuild/RPMS/ rpmbuild/RPMS/SHA256SUMS rpmbuild/RPMS/SHA256SUMS.as
rsync -avz --prune-empty-dirs rpmbuild/RPMS/ $(SERVER):$(REMOTE_PATH)/RPMS/; \
fi
upload-releases:
echo Running releases/ upload...
rsync -avz --exclude '.mkdir' --exclude 'mgmt-release.url' releases/ $(SERVER):$(REMOTE_PATH)/releases/
#
# copr build
#
copr: upload-srpms
copr: upload-srpms ## build in copr
./misc/copr-build.py https://$(SERVER)/$(REMOTE_PATH)/SRPMS/$(SRPM_BASE)
#
# tag
#
tag: ## tags a new release
./misc/tag.sh
#
# mkosi
#
mkosi: mkosi_fedora-latest mkosi_fedora-older mkosi_stream-latest mkosi_debian-stable mkosi_ubuntu-latest mkosi_archlinux ## builds distro packages via mkosi
mkosi_fedora-latest: releases/$(VERSION)/.mkdir
@title='$@' ; echo "Generating: $${title#'mkosi_'} via mkosi..."
@title='$@' ; distro=$${title#'mkosi_'} ; ./misc/mkosi/make.sh $${distro} `realpath "releases/$(VERSION)/"`
mkosi_fedora-older: releases/$(VERSION)/.mkdir
@title='$@' ; echo "Generating: $${title#'mkosi_'} via mkosi..."
@title='$@' ; distro=$${title#'mkosi_'} ; ./misc/mkosi/make.sh $${distro} `realpath "releases/$(VERSION)/"`
mkosi_stream-latest: releases/$(VERSION)/.mkdir
@title='$@' ; echo "Generating: $${title#'mkosi_'} via mkosi..."
@title='$@' ; distro=$${title#'mkosi_'} ; ./misc/mkosi/make.sh $${distro} `realpath "releases/$(VERSION)/"`
mkosi_debian-stable: releases/$(VERSION)/.mkdir
@title='$@' ; echo "Generating: $${title#'mkosi_'} via mkosi..."
@title='$@' ; distro=$${title#'mkosi_'} ; ./misc/mkosi/make.sh $${distro} `realpath "releases/$(VERSION)/"`
mkosi_ubuntu-latest: releases/$(VERSION)/.mkdir
@title='$@' ; echo "Generating: $${title#'mkosi_'} via mkosi..."
@title='$@' ; distro=$${title#'mkosi_'} ; ./misc/mkosi/make.sh $${distro} `realpath "releases/$(VERSION)/"`
mkosi_archlinux: releases/$(VERSION)/.mkdir
@title='$@' ; echo "Generating: $${title#'mkosi_'} via mkosi..."
@title='$@' ; distro=$${title#'mkosi_'} ; ./misc/mkosi/make.sh $${distro} `realpath "releases/$(VERSION)/"`
#
# release
#
release: TRIMPATH = -trimpath
release: releases/$(VERSION)/mgmt-release.url ## generates and uploads a release
releases_path:
@#Don't put any other output or dependencies in here or they'll show!
@echo "releases/$(VERSION)/"
release_test: $(DEP_BINARY_AMD64) $(DEP_BINARY_ARM64) $(DEP_FEDORA-LATEST) $(DEP_FEDORA-OLDER) $(DEP_STREAM-LATEST) $(DEP_DEBIAN-STABLE) $(DEP_UBUNTU-LATEST) $(DEP_ARCHLINUX) $(SHA256SUMS_ASC)
@echo '$$< denotes the first dependency of the current rule.'
@echo '> '"$<"
@echo
@echo '$$@ denotes the target of the current rule.'
@echo '> '"$@"
@echo
@echo '$$^ denotes the dependencies of the current rule.'
@echo '> '"$^"
@echo
@echo '$$* denotes the stem with which the pattern of the current rule matched.'
@echo '> '"$*"
@echo
@echo "TOKEN_BINARY_AMD64: $(TOKEN_BINARY_AMD64)"
@echo "DEP_BINARY_AMD64: $(DEP_BINARY_AMD64)"
@echo
@echo "TOKEN_BINARY_ARM64: $(TOKEN_BINARY_ARM64)"
@echo "DEP_BINARY_ARM64: $(DEP_BINARY_ARM64)"
@echo
@echo "TOKEN_FEDORA-LATEST: $(TOKEN_FEDORA-LATEST)"
@echo "DEP_FEDORA-LATEST: $(DEP_FEDORA-LATEST)"
@echo
@echo "TOKEN_FEDORA-OLDER: $(TOKEN_FEDORA-OLDER)"
@echo "DEP_FEDORA-OLDER: $(DEP_FEDORA-OLDER)"
@echo
@echo "TOKEN_STREAM-LATEST: $(TOKEN_STREAM-LATEST)"
@echo "DEP_STREAM-LATEST: $(DEP_STREAM-LATEST)"
@echo
@echo "TOKEN_DEBIAN-STABLE: $(TOKEN_DEBIAN-STABLE)"
@echo "DEP_DEBIAN-STABLE: $(DEP_DEBIAN-STABLE)"
@echo
@echo "TOKEN_UBUNTU-LATEST: $(TOKEN_UBUNTU-LATEST)"
@echo "DEP_UBUNTU-LATEST: $(DEP_UBUNTU-LATEST)"
@echo
@echo "TOKEN_ARCHLINUX: $(TOKEN_ARCHLINUX)"
@echo "DEP_ARCHLINUX: $(DEP_ARCHLINUX)"
release_binary_amd64: $(PKG_BINARY_AMD64)
release_binary_arm64: $(PKG_BINARY_ARM64)
release_fedora-latest: $(PKG_FEDORA-LATEST)
release_fedora-older: $(PKG_FEDORA-OLDER)
release_stream-latest: $(PKG_STREAM-LATEST)
release_debian-stable: $(PKG_DEBIAN-STABLE)
release_ubuntu-latest: $(PKG_UBUNTU-LATEST)
release_archlinux: $(PKG_ARCHLINUX)
releases/$(VERSION)/mgmt-release.url: $(DEP_BINARY_AMD64) $(DEP_BINARY_ARM64) $(DEP_FEDORA-LATEST) $(DEP_FEDORA-OLDER) $(DEP_STREAM-LATEST) $(DEP_DEBIAN-STABLE) $(DEP_UBUNTU-LATEST) $(DEP_ARCHLINUX) $(SHA256SUMS_ASC)
@echo "Pushing git tag $(VERSION) to origin..."
git push origin $(VERSION)
@echo "Creating github release..."
hub release create \
-F <( echo -e "$(VERSION)\n";echo "Verify the signatures of all packages before you use them. The signing key can be downloaded from https://purpleidea.com/contact/#pgp-key to verify the release." ) \
` [ -e $(PKG_BINARY_AMD64) ] && printf -- "-a $(PKG_BINARY_AMD64)" ` \
` [ -e $(PKG_BINARY_ARM64) ] && printf -- "-a $(PKG_BINARY_ARM64)" ` \
` [ -e $(PKG_FEDORA-LATEST) ] && printf -- "-a $(PKG_FEDORA-LATEST)" ` \
` [ -e $(PKG_FEDORA-OLDER) ] && printf -- "-a $(PKG_FEDORA-OLDER)" ` \
` [ -e $(PKG_STREAM-LATEST) ] && printf -- "-a $(PKG_STREAM-LATEST)" ` \
` [ -e $(PKG_DEBIAN-STABLE) ] && printf -- "-a $(PKG_DEBIAN-STABLE)" ` \
` [ -e $(PKG_UBUNTU-LATEST) ] && printf -- "-a $(PKG_UBUNTU-LATEST)" ` \
` [ -e $(PKG_ARCHLINUX) ] && printf -- "-a $(PKG_ARCHLINUX)" ` \
-a $(SHA256SUMS_ASC) \
$(VERSION) \
> releases/$(VERSION)/mgmt-release.url \
&& cat releases/$(VERSION)/mgmt-release.url \
|| rm -f releases/$(VERSION)/mgmt-release.url
releases/$(VERSION)/.mkdir:
mkdir -p \
` [ "$(TOKEN_BINARY_AMD64)" != "" ] && printf -- "releases/$(VERSION)/$(TOKEN_BINARY_AMD64)/" ` \
` [ "$(TOKEN_BINARY_ARM64)" != "" ] && printf -- "releases/$(VERSION)/$(TOKEN_BINARY_ARM64)/" ` \
` [ "$(TOKEN_FEDORA-LATEST)" != "" ] && printf -- "releases/$(VERSION)/$(TOKEN_FEDORA-LATEST)/" ` \
` [ "$(TOKEN_FEDORA-OLDER)" != "" ] && printf -- "releases/$(VERSION)/$(TOKEN_FEDORA-OLDER)/" ` \
` [ "$(TOKEN_STREAM-LATEST)" != "" ] && printf -- "releases/$(VERSION)/$(TOKEN_STREAM-LATEST)/" ` \
` [ "$(TOKEN_DEBIAN-STABLE)" != "" ] && printf -- "releases/$(VERSION)/$(TOKEN_DEBIAN-STABLE)/" ` \
` [ "$(TOKEN_UBUNTU-LATEST)" != "" ] && printf -- "releases/$(VERSION)/$(TOKEN_UBUNTU-LATEST)/" ` \
` [ "$(TOKEN_ARCHLINUX)" != "" ] && printf -- "releases/$(VERSION)/$(TOKEN_ARCHLINUX)/" ` \
&& touch releases/$(VERSION)/.mkdir
# These are defined conditionally, since if the token is empty, they warn!
ifneq ($(TOKEN_BINARY_AMD64),)
$(PKG_BINARY_AMD64): build/mgmt-linux-amd64 releases/$(VERSION)/.mkdir
@title='$(@D)' ; distro=$${title#'releases/$(VERSION)/'} ; echo "Building: $${distro} package..."
@title='$(@D)' ; distro=$${title#'releases/$(VERSION)/'} ; cp -a build/mgmt-linux-amd64 $(PKG_BINARY_AMD64)
endif
ifneq ($(TOKEN_BINARY_ARM64),)
$(PKG_BINARY_ARM64): build/mgmt-linux-arm64 releases/$(VERSION)/.mkdir
@title='$(@D)' ; distro=$${title#'releases/$(VERSION)/'} ; echo "Building: $${distro} package..."
@title='$(@D)' ; distro=$${title#'releases/$(VERSION)/'} ; cp -a build/mgmt-linux-arm64 $(PKG_BINARY_ARM64)
endif
ifneq ($(TOKEN_FEDORA-LATEST),)
releases/$(VERSION)/$(TOKEN_FEDORA-LATEST)/changelog: $(PROGRAM) releases/$(VERSION)/.mkdir
@title='$(@D)' ; distro=$${title#'releases/$(VERSION)/'} ; echo "Generating: $${distro} changelog..."
@title='$(@D)' ; distro=$${title#'releases/$(VERSION)/'} ; ./misc/make-rpm-changelog.sh "$${distro}" $(VERSION)
$(PKG_FEDORA-LATEST): releases/$(VERSION)/$(TOKEN_FEDORA-LATEST)/changelog
@title='$(@D)' ; distro=$${title#'releases/$(VERSION)/'} ; echo "Building: $${distro} package..."
@title='$(@D)' ; distro=$${title#'releases/$(VERSION)/'} ; ./misc/fpm-pack.sh $${distro} $(VERSION) "$(FILE_FEDORA-LATEST)" libvirt-devel augeas-devel
endif
ifneq ($(TOKEN_FEDORA-OLDER),)
releases/$(VERSION)/$(TOKEN_FEDORA-OLDER)/changelog: $(PROGRAM) releases/$(VERSION)/.mkdir
@title='$(@D)' ; distro=$${title#'releases/$(VERSION)/'} ; echo "Generating: $${distro} changelog..."
@title='$(@D)' ; distro=$${title#'releases/$(VERSION)/'} ; ./misc/make-rpm-changelog.sh "$${distro}" $(VERSION)
$(PKG_FEDORA-OLDER): releases/$(VERSION)/$(TOKEN_FEDORA-OLDER)/changelog
@title='$(@D)' ; distro=$${title#'releases/$(VERSION)/'} ; echo "Building: $${distro} package..."
@title='$(@D)' ; distro=$${title#'releases/$(VERSION)/'} ; ./misc/fpm-pack.sh $${distro} $(VERSION) "$(FILE_FEDORA-OLDER)" libvirt-devel augeas-devel
endif
ifneq ($(TOKEN_STREAM-LATEST),)
releases/$(VERSION)/$(TOKEN_STREAM-LATEST)/changelog: $(PROGRAM) releases/$(VERSION)/.mkdir
@title='$(@D)' ; distro=$${title#'releases/$(VERSION)/'} ; echo "Generating: $${distro} changelog..."
@title='$(@D)' ; distro=$${title#'releases/$(VERSION)/'} ; ./misc/make-rpm-changelog.sh "$${distro}" $(VERSION)
$(PKG_STREAM-LATEST): releases/$(VERSION)/$(TOKEN_STREAM-LATEST)/changelog
@title='$(@D)' ; distro=$${title#'releases/$(VERSION)/'} ; echo "Building: $${distro} package..."
@title='$(@D)' ; distro=$${title#'releases/$(VERSION)/'} ; ./misc/fpm-pack.sh $${distro} $(VERSION) "$(FILE_STREAM-LATEST)" libvirt-devel augeas-devel
endif
ifneq ($(TOKEN_DEBIAN-STABLE),)
releases/$(VERSION)/$(TOKEN_DEBIAN-STABLE)/changelog: $(PROGRAM) releases/$(VERSION)/.mkdir
@title='$(@D)' ; distro=$${title#'releases/$(VERSION)/'} ; echo "Generating: $${distro} changelog..."
@title='$(@D)' ; distro=$${title#'releases/$(VERSION)/'} ; ./misc/make-deb-changelog.sh "$${distro}" $(VERSION)
$(PKG_DEBIAN-STABLE): releases/$(VERSION)/$(TOKEN_DEBIAN-STABLE)/changelog
@title='$(@D)' ; distro=$${title#'releases/$(VERSION)/'} ; echo "Building: $${distro} package..."
@title='$(@D)' ; distro=$${title#'releases/$(VERSION)/'} ; ./misc/fpm-pack.sh $${distro} $(VERSION) "$(FILE_DEBIAN-STABLE)" libvirt-dev libaugeas-dev
endif
ifneq ($(TOKEN_UBUNTU-LATEST),)
releases/$(VERSION)/$(TOKEN_UBUNTU-LATEST)/changelog: $(PROGRAM) releases/$(VERSION)/.mkdir
@title='$(@D)' ; distro=$${title#'releases/$(VERSION)/'} ; echo "Generating: $${distro} changelog..."
@title='$(@D)' ; distro=$${title#'releases/$(VERSION)/'} ; ./misc/make-deb-changelog.sh "$${distro}" $(VERSION)
$(PKG_UBUNTU-LATEST): releases/$(VERSION)/$(TOKEN_UBUNTU-LATEST)/changelog
@title='$(@D)' ; distro=$${title#'releases/$(VERSION)/'} ; echo "Building: $${distro} package..."
@title='$(@D)' ; distro=$${title#'releases/$(VERSION)/'} ; ./misc/fpm-pack.sh $${distro} $(VERSION) "$(FILE_UBUNTU-LATEST)" libvirt-dev libaugeas-dev
endif
ifneq ($(TOKEN_ARCHLINUX),)
$(PKG_ARCHLINUX): $(PROGRAM) releases/$(VERSION)/.mkdir
@title='$(@D)' ; distro=$${title#'releases/$(VERSION)/'} ; echo "Building: $${distro} package..."
@title='$(@D)' ; distro=$${title#'releases/$(VERSION)/'} ; ./misc/fpm-pack.sh $${distro} $(VERSION) "$(FILE_ARCHLINUX)" libvirt augeas
endif
$(SHA256SUMS): $(DEP_BINARY_AMD64) $(DEP_BINARY_ARM64) $(DEP_FEDORA-LATEST) $(DEP_FEDORA-OLDER) $(DEP_STREAM-LATEST) $(DEP_DEBIAN-STABLE) $(DEP_UBUNTU-LATEST) $(DEP_ARCHLINUX)
@# remove the directory separator in the SHA256SUMS file
@echo "Generating: sha256 sum..."
sha256sum \
` [ -e $(PKG_BINARY_AMD64) ] && printf -- "$(PKG_BINARY_AMD64)" ` \
` [ -e $(PKG_BINARY_ARM64) ] && printf -- "$(PKG_BINARY_ARM64)" ` \
` [ -e $(PKG_FEDORA-LATEST) ] && printf -- "$(PKG_FEDORA-LATEST)" ` \
` [ -e $(PKG_FEDORA-OLDER) ] && printf -- "$(PKG_FEDORA-OLDER)" ` \
` [ -e $(PKG_STREAM-LATEST) ] && printf -- "$(PKG_STREAM-LATEST)" ` \
` [ -e $(PKG_DEBIAN-STABLE) ] && printf -- "$(PKG_DEBIAN-STABLE)" ` \
` [ -e $(PKG_UBUNTU-LATEST) ] && printf -- "$(PKG_UBUNTU-LATEST)" ` \
` [ -e $(PKG_ARCHLINUX) ] && printf -- "$(PKG_ARCHLINUX)" ` \
| awk -F '/| ' '{print $$1" "$$6}' > $(SHA256SUMS)
$(SHA256SUMS_ASC): $(SHA256SUMS)
@echo "Signing sha256 sum..."
gpg2 --yes --clearsign $(SHA256SUMS)
build_container: ## builds the container
docker build -t purpleidea/mgmt-build -f docker/Dockerfile.build .
docker run -td --name mgmt-build purpleidea/mgmt-build
docker cp mgmt-build:/root/gopath/src/github.com/purpleidea/mgmt/mgmt .
docker build -t purpleidea/mgmt -f docker/Dockerfile.static .
docker rm mgmt-build || true
clean_container: ## removes the container
docker rmi purpleidea/mgmt-build
docker rmi purpleidea/mgmt
help: ## show this help screen
@echo 'Usage: make <OPTIONS> ... <TARGETS>'
@echo ''
@echo 'Available targets are:'
@echo ''
@grep -E '^[ a-zA-Z_-]+:.*?## .*$$' $(MAKEFILE_LIST) | \
awk 'BEGIN {FS = ":.*?## "}; {printf "\033[36m%-20s\033[0m %s\n", $$1, $$2}'
@echo ''
funcgen: lang/core/generated_funcs.go
lang/core/generated_funcs.go: lang/funcs/funcgen/*.go lang/core/funcgen.yaml lang/funcs/funcgen/templates/generated_funcs.go.tpl
@echo "Generating: funcs..."
@go run `find lang/funcs/funcgen/ -maxdepth 1 -type f -name '*.go' -not -name '*_test.go'` -templates=lang/funcs/funcgen/templates/generated_funcs.go.tpl >/dev/null
@gofmt -s -w $@
# vim: ts=8

179
README.md
View File

@@ -1,66 +1,153 @@
# *mgmt*: This is: mgmt!
# *mgmt*: next generation config management!
[![Build Status](https://secure.travis-ci.org/purpleidea/mgmt.png?branch=master)](http://travis-ci.org/purpleidea/mgmt)
[![Documentation](https://img.shields.io/docs/markdown.png)](DOCUMENTATION.md)
[![IRC](https://img.shields.io/irc/%23mgmtconfig.png)](https://webchat.freenode.net/?channels=#mgmtconfig)
[![Jenkins](https://img.shields.io/jenkins/status.png)](https://ci.centos.org/job/purpleidea-mgmt/)
[![COPR](https://img.shields.io/copr/builds.png)](https://copr.fedoraproject.org/coprs/purpleidea/mgmt/)
[![mgmt!](art/mgmt.png)](art/)
[![Go Report Card](https://goreportcard.com/badge/github.com/purpleidea/mgmt?style=flat-square)](https://goreportcard.com/report/github.com/purpleidea/mgmt)
[![Build Status](https://github.com/purpleidea/mgmt/workflows/.github/workflows/test.yaml/badge.svg)](https://github.com/purpleidea/mgmt/actions/)
[![GoDoc](https://img.shields.io/badge/godoc-reference-5272B4.svg?style=flat-square)](https://godocs.io/github.com/purpleidea/mgmt)
[![Matrix](https://img.shields.io/badge/matrix-%23mgmtconfig-orange.svg?style=flat-square)](https://matrix.to/#/#mgmtconfig:matrix.org)
[![Patreon](https://img.shields.io/badge/patreon-donate-yellow.svg?style=flat-square)](https://www.patreon.com/purpleidea)
[![Liberapay](https://img.shields.io/badge/liberapay-donate-yellow.svg?style=flat-square)](https://liberapay.com/purpleidea/donate)
> [!TIP]
> [Resource reference guide now available!](https://mgmtconfig.com/docs/resources/)
> [!TIP]
> [Function reference guide now available!](https://mgmtconfig.com/docs/functions/)
## About:
`Mgmt` is a real-time automation tool. It is familiar to existing configuration
management software, but is drastically more powerful as it can allow you to
build real-time, closed-loop feedback systems, in a very safe way, and with a
surprisingly small amount of our `mcl` code. For example, the following code
will ensure that your file server is set to read-only when it's friday.
```mcl
import "datetime"
$is_friday = datetime.weekday(datetime.now()) == "friday"
file "/srv/files/" {
state => $const.res.file.state.exists,
mode => if $is_friday { # this updates the mode, the instant it changes!
"0550"
} else {
"0770"
},
}
```
It can run continuously, intermittently, or on-demand, and in the first case, it
will guarantee that your system is always in the desired state for that instant!
In this mode it can run as a decentralized cluster of agents across your
network, each exchanging information with the others in real-time, to respond to
your changing needs. For example, if you want to ensure that some resource runs
on a maximum of two hosts in your cluster, you can specify that as well:
```mcl
import "sys"
import "world"
# we'll set a few scheduling options:
$opts = struct{strategy => "rr", max => 2, ttl => 10,}
# schedule in a particular namespace with options:
$set = world.schedule("xsched", $opts)
if sys.hostname() in $set {
# use your imagination to put something more complex right here...
print "i got scheduled" {} # this will run on the chosen machines
}
```
As you add and remove hosts from the cluster, the real-time `schedule` function
will dynamically pick up to two hosts from the available pool. These specific
functions aren't intrinsic to the core design, and new ones can be easily added.
Please read on if you'd like to learn more...
## Community:
Come join us on IRC in [#mgmtconfig](https://webchat.freenode.net/?channels=#mgmtconfig) on Freenode!
You may like the [#mgmtconfig](https://twitter.com/hashtag/mgmtconfig) hashtag if you're on [Twitter](https://twitter.com/#!/purpleidea).
## Questions:
Please join the [#mgmtconfig](https://webchat.freenode.net/?channels=#mgmtconfig) IRC community!
If you have a well phrased question that might benefit others, consider asking it by sending a patch to the documentation [FAQ](https://github.com/purpleidea/mgmt/blob/master/DOCUMENTATION.md#usage-and-frequently-asked-questions) section. I'll merge your question, and a patch with the answer!
Come join us in the `mgmt` community!
## Quick start:
* Either get the golang dependencies on your own, or run `make deps` if you're comfortable with how we install them.
* Run `make build` to get a fresh built `mgmt` binary.
* Run `cd $(mktemp --tmpdir -d tmp.XXX) && etcd` to get etcd running. The `mgmt` software will do this automatically for you in the future.
* Run `time ./mgmt run --file examples/graph0.yaml --converged-timeout=1` to try out a very simple example!
* To run continuously in the default mode of operation, omit the `--converged-timeout` option.
* Have fun hacking on our future technology!
| Medium | Link |
|---|---|
| Matrix | [#mgmtconfig](https://matrix.to/#/#mgmtconfig:matrix.org) on Matrix.org |
| Twitter | [@mgmtconfig](https://twitter.com/mgmtconfig) & [#mgmtconfig](https://twitter.com/hashtag/mgmtconfig) |
| Mailing list | [looking for a new home, suggestions welcome](https://gitlab.freedesktop.org/freedesktop/freedesktop/-/issues/1082) |
| Patreon | [purpleidea](https://www.patreon.com/purpleidea) on Patreon |
## Examples:
Please look in the [examples/](examples/) folder for more examples!
## Status:
Mgmt is a next generation automation tool. It has similarities to other tools in
the configuration management space, but has a fast, modern, distributed systems
approach. The project contains an engine and a language.
[Please have a look at an introductory video or blog post.](docs/on-the-web.md)
Mgmt is over ten years old! It is very powerful today, and has a solid
foundation and architecture which has been polished over the years. As with all
software, there are bugs to fix and improvements to be made, but I expect
they're easy to hack through and fix if you find any. Interested users should
start with the [official website](https://mgmtconfig.com/docs/).
## Sponsors:
Mgmt is generously sponsored by:
[![m9rx corporation](art/m9rx.png)](https://m9rx.com/)
Please reach out if you'd like to sponsor!
## Documentation:
Please see: [DOCUMENTATION.md](DOCUMENTATION.md) or [PDF](https://pdfdoc-purpleidea.rhcloud.com/pdf/https://github.com/purpleidea/mgmt/blob/master/DOCUMENTATION.md).
## Roadmap:
Please see: [TODO.md](TODO.md) for a list of upcoming work and TODO items.
Please get involved by working on one of these items or by suggesting something else!
Please read, enjoy and help improve our documentation!
| Documentation | Additional Notes |
|---|---|
| [quick start guide](docs/quick-start-guide.md) | for everyone |
| [frequently asked questions](docs/faq.md) | for everyone |
| [general documentation](docs/documentation.md) | for everyone |
| [resource reference](https://mgmtconfig.com/docs/resources/) | for everyone |
| [function reference](https://mgmtconfig.com/docs/functions/) | for everyone |
| [language guide](docs/language-guide.md) | for everyone |
| [function guide](docs/function-guide.md) | for mgmt developers |
| [resource guide](docs/resource-guide.md) | for mgmt developers |
| [style guide](docs/style-guide.md) | for mgmt developers |
| [contributing guide](docs/contributing.md) | for mgmt contributors |
| [service API guide](docs/service-guide.md) | for external developers |
| [godoc API reference](https://godoc.org/github.com/purpleidea/mgmt) | for mgmt developers |
| [prometheus guide](docs/prometheus.md) | for everyone |
| [puppet guide](docs/puppet-guide.md) | for puppet sysadmins |
| [development](docs/development.md) | for mgmt developers |
| [videos](docs/on-the-web.md) | for everyone |
| [blogs](docs/on-the-web.md) | for everyone |
## Questions:
Please ask in the [community](#community)!
If you have a well phrased question that might benefit others, consider asking
it by sending a patch to the [FAQ](docs/faq.md) section. I'll merge your
question, and a patch with the answer!
## Get involved:
Feel free to grab one of the straightforward [#mgmtlove](https://github.com/purpleidea/mgmt/labels/mgmtlove)
issues if you're a first time contributor to the project or if you're unsure
about what to hack on! Please get involved by working on one of these items or
by suggesting something else! There are some lower priority issues and harder
issues available in our [TODO](TODO.md) file. Please have a look.
## Bugs:
Please set the `DEBUG` constant in [main.go](https://github.com/purpleidea/mgmt/blob/master/main.go) to `true`, and post the logs when you report the [issue](https://github.com/purpleidea/mgmt/issues).
Bonus points if you provide a [shell](https://github.com/purpleidea/mgmt/tree/master/test/shell) or [OMV](https://github.com/purpleidea/mgmt/tree/master/test/omv) reproducible test case.
## Dependencies:
* golang 1.4 or higher (required, available in most distros)
* golang libraries (required, available with `go get`)
go get github.com/coreos/etcd/client
go get gopkg.in/yaml.v2
go get gopkg.in/fsnotify.v1
go get github.com/codegangsta/cli
go get github.com/coreos/go-systemd/dbus
go get github.com/coreos/go-systemd/util
* stringer (required for building), available as a package on some platforms, otherwise via `go get`
go get golang.org/x/tools/cmd/stringer
* pandoc (optional, for building a pdf of the documentation)
* graphviz (optional, for building a visual representation of the graph)
Please set the `DEBUG` constant in [main.go](https://github.com/purpleidea/mgmt/blob/master/main.go)
to `true`, and post the logs when you report the [issue](https://github.com/purpleidea/mgmt/issues).
Feel free to read my article on [debugging golang programs](https://purpleidea.com/blog/2016/02/15/debugging-golang-programs/).
## Patches:
We'd love to have your patches! Please send them by email, or as a pull request.
## On the web:
* Introductory blog post: [https://ttboj.wordpress.com/2016/01/18/next-generation-configuration-mgmt/](https://ttboj.wordpress.com/2016/01/18/next-generation-configuration-mgmt/)
* Julian Dunn at Cfgmgmtcamp 2016 [https://www.youtube.com/watch?v=kfF9IATUask&t=1949&html5=1](https://www.youtube.com/watch?v=kfF9IATUask&t=1949&html5=1)
##
[Blog posts and recorded talks about mgmt are listed here!](docs/on-the-web.md)
Happy hacking!

6
THANKS
View File

@@ -9,10 +9,16 @@ Chris Wright - For encouraging me to continue work on my prototype.
Daniel Riek - For supporting and sheltering this project from bureaucracy.
Diego Ongaro - For good chats, particularly around distributed systems.
Felix Frank - For taking a difficult problem and building an inspiring solution.
Ira Cooper - For having an algorithmic design discussion with me.
Jeff Darcy - For some algorithm recommendations, and NACKing my TopoSort idea!
Red Hat, inc. - For paying my salary, thus financially supporting my hacking.
Samuel Gélineau - For help with programming language theory and design.
And many others...

91
TODO.md
View File

@@ -1,43 +1,90 @@
# TODO
If you're looking for something to do, look here!
Let us know if you're working on one of the items.
Here is a TODO list of longstanding items that are either lower-priority, or
more involved in terms of time, skill-level, and/or motivation.
Please have a look, and let us know if you're working on one of the items. It's
best to open an issue to track your progress and to discuss any implementation
questions you might have.
Lastly, if you'd like something different to work on, please ping @purpleidea
and I'll create an issue tailored especially for your approximate golang skill
level and available time commitment in terms of hours you'd need to spend on the
patch.
Happy Hacking!
## Package resource
- [ ] base type [bug](https://github.com/purpleidea/mgmt/issues/11)
- [ ] dnf blocker [bug](https://github.com/hughsie/PackageKit/issues/110)
- [ ] install signal blocker [bug](https://github.com/hughsie/PackageKit/issues/109)
## File resource
- [ ] ability to make/delete folders
- [ ] recursive argument (can recursively watch/modify contents)
- [ ] force argument (can cause switch from file <-> folder)
- [ ] getfiles support on debian [bug](https://github.com/hughsie/PackageKit/issues/118)
- [ ] directory info on fedora [bug](https://github.com/hughsie/PackageKit/issues/117)
- [ ] dnf blocker [bug](https://github.com/hughsie/PackageKit/issues/110)
## File resource [bug](https://github.com/purpleidea/mgmt/issues/64) [:heart:](https://github.com/purpleidea/mgmt/labels/mgmtlove)
- [ ] recurse limit support [:heart:](https://github.com/purpleidea/mgmt/labels/mgmtlove)
- [ ] fanotify support [bug](https://github.com/go-fsnotify/fsnotify/issues/114)
## Svc resource
- [ ] refreshonly support [:heart:](https://github.com/purpleidea/mgmt/issues/464)
## Exec resource
- [ ] base resource improvements
## Timer resource
- [ ] base resource
- [ ] reset on recompile
- [ ] increment algorithm (linear, exponential, etc...)
- [ ] increment algorithm (linear, exponential, etc...) [:heart:](https://github.com/purpleidea/mgmt/labels/mgmtlove)
## User/Group resource
- [ ] automatic edges to file resource [:heart:](https://github.com/purpleidea/mgmt/labels/mgmtlove)
## Http resource
- [ ] base resource [:heart:](https://github.com/purpleidea/mgmt/labels/mgmtlove)
## Etcd improvements
- [ ] embedded etcd master
- [ ] capnslog fixes [bug](https://github.com/coreos/etcd/issues/4115)
- [ ] fix etcd race bug that only happens during CI testing (intermittently
failing test case issue)
## Torrent/dht file transfer
- [ ] base plumbing
## GPG/Auth improvements
- [ ] base plumbing
## Resource improvements
- [ ] more reversible resources implemented
- [ ] more "cloud" resources
## Language improvements
- [ ] language design
- [ ] lexer/parser
- [ ] more core functions
- [ ] automatic language formatter, ala `gofmt`
- [ ] gedit/gnome-builder/gtksourceview syntax highlighting
- [ ] vim syntax highlighting
- [ ] emacs syntax highlighting
- [ ] emacs syntax highlighting: see `misc/emacs/` (needs updating)
- [ ] exposed $error variable for feedback in the language
- [ ] improve the printf function to add %[]s, %[]f ([]str, []float) and map,
struct, nested etc... %v would be nice too!
- [ ] add line/col/file annotations to AST so we can get locations of errors
that the parser finds
- [ ] add more error messages with the `%error` pattern in parser.y
- [ ] we should have helper functions or language sugar to pull a field out of a
struct, or a value out of a map, or an index out of a list, etc...
## Engine improvements
- [ ] add a "waiting for func" message in the func engine to notify the user
about slow functions...
## Other
- [ ] better error/retry handling
- [ ] resource grouping
- [ ] automatic dependency adding (eg: packagekit file dependencies)
- [ ] rpm package target in Makefile
- [ ] deb package target in Makefile
- [ ] reproducible builds
- [ ] add your suggestions!

51
Vagrantfile vendored Normal file
View File

@@ -0,0 +1,51 @@
Vagrant.configure(2) do |config|
config.ssh.forward_agent = true
config.ssh.username = 'vagrant'
config.vm.network "private_network", ip: "192.168.219.2"
config.vm.synced_folder ".", "/vagrant", disabled: true
config.vm.define "mgmt-dev" do |instance|
instance.vm.box = "bento/fedora-31"
end
config.vm.provider "virtualbox" do |v|
v.memory = 1536
v.cpus = 2
end
config.vm.provider "libvirt" do |v|
v.memory = 2048
end
config.vm.provision "file", source: "vagrant/motd", destination: ".motd"
config.vm.provision "shell", inline: "cp ~vagrant/.motd /etc/motd"
config.vm.provision "file", source: "vagrant/mgmt.bashrc", destination: ".mgmt.bashrc"
config.vm.provision "file", source: "~/.gitconfig", destination: ".gitconfig"
config.vm.provision "shell", inline: "dnf install -y golang git make"
# set up packagekit
config.vm.provision "shell" do |shell|
shell.inline = <<-SCRIPT
dnf install -y PackageKit
systemctl enable packagekit
systemctl start packagekit
SCRIPT
end
# set up vagrant home
script = <<-SCRIPT
grep -q 'mgmt\.bashrc' ~/.bashrc || echo '. ~/.mgmt.bashrc' >>~/.bashrc
. ~/.mgmt.bashrc
mkdir -p ~/gopath/src/github.com/purpleidea
cd ~/gopath/src/github.com/purpleidea
git clone https://github.com/purpleidea/mgmt --recursive
cd mgmt
make deps
SCRIPT
config.vm.provision "shell" do |shell|
shell.privileged = false
shell.inline = script
end
end

2
art/.gitignore vendored Normal file
View File

@@ -0,0 +1,2 @@
*.png
misc/

BIN
art/m9rx.png Normal file

Binary file not shown.

After

Width:  |  Height:  |  Size: 37 KiB

BIN
art/mgmt.png Normal file

Binary file not shown.

After

Width:  |  Height:  |  Size: 683 KiB

View File

@@ -0,0 +1,94 @@
<?xml version="1.0" encoding="utf-8"?>
<svg version="1.1" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" x="0px" y="0px"
viewBox="0 0 120 107.1" style="enable-background:new 0 0 120 107.1;" xml:space="preserve">
<style type="text/css">
.st0{fill:#FFFFFF;}
.st1{fill:#E22434;}
.st2{display:none;}
.st3{fill:#1B3663;}
.st4{fill:#00B1D1;}
.st5{fill:#BFE6EF;}
.st6{fill:#69CBE0;}
.st7{fill:#0080BD;}
.st8{fill:#005DAB;}
.st9{fill:#183660;}
.st10{fill:none;stroke:#183660;stroke-width:1.9441;stroke-miterlimit:10;}
.st11{fill:none;stroke:#183660;stroke-width:1.6201;stroke-miterlimit:10;}
.st12{fill:none;stroke:#183660;stroke-width:1.2961;stroke-miterlimit:10;}
.st13{fill:none;stroke:#C0E6EF;stroke-width:1.9441;stroke-miterlimit:10;}
.st14{fill:#C0E6EF;}
.st15{fill:none;stroke:#C0E6EF;stroke-width:1.2961;stroke-miterlimit:10;}
.st16{fill:none;stroke:#C0E6EF;stroke-width:1.6201;stroke-miterlimit:10;}
.st17{fill:none;stroke:#FFFFFF;stroke-width:1.9441;stroke-miterlimit:10;}
.st18{fill:none;stroke:#FFFFFF;stroke-width:1.2961;stroke-miterlimit:10;}
.st19{fill:none;stroke:#FFFFFF;stroke-width:1.6201;stroke-miterlimit:10;}
</style>
<g id="Layer_2">
</g>
<g id="Layer_1">
<g>
<g>
<g>
<line class="st10" x1="29.2" y1="24.1" x2="52.1" y2="42.8"/>
<g>
<polygon class="st9" points="27.7,27.8 24.9,20.5 32.6,21.8 "/>
</g>
</g>
</g>
<circle class="st5" cx="16.1" cy="12.2" r="12.1"/>
<g>
<g>
<line class="st10" x1="52.1" y1="42.1" x2="74.1" y2="80"/>
<g>
<polygon class="st9" points="70.1,80.9 76.9,84.8 76.8,77.1 "/>
</g>
</g>
</g>
<g>
<g>
<line class="st10" x1="69.4" y1="46.7" x2="95.8" y2="52.4"/>
<g>
<polygon class="st9" points="69.7,50.7 63.9,45.6 71.3,43.2 "/>
</g>
</g>
</g>
<g>
<g>
<line class="st10" x1="52.1" y1="42.8" x2="71.9" y2="30.1"/>
<g>
<polygon class="st9" points="73.1,34 76.6,27.1 68.9,27.5 "/>
</g>
</g>
</g>
<g>
<g>
<line class="st10" x1="16.8" y1="49.6" x2="34.8" y2="46.5"/>
<g>
<polygon class="st9" points="34.3,50.5 40.3,45.6 33,42.9 "/>
</g>
</g>
</g>
<g>
<g>
<line class="st10" x1="92.3" y1="79.5" x2="107.8" y2="54.2"/>
<g>
<polygon class="st9" points="96.1,80.6 89.3,84.3 89.5,76.5 "/>
</g>
</g>
</g>
<g>
<g>
<line class="st10" x1="97.3" y1="36.5" x2="107.8" y2="54.2"/>
<g>
<polygon class="st9" points="94.6,39.4 94.5,31.7 101.2,35.5 "/>
</g>
</g>
</g>
<circle class="st3" cx="52.1" cy="42.8" r="12.1"/>
<circle class="st4" cx="12.2" cy="50.8" r="12.1"/>
<circle class="st7" cx="87.5" cy="21.7" r="12.1"/>
<circle class="st8" cx="83.5" cy="95" r="12.1"/>
<circle class="st6" cx="107.8" cy="54.2" r="12.1"/>
</g>
</g>
</svg>

After

Width:  |  Height:  |  Size: 2.8 KiB

View File

@@ -0,0 +1,132 @@
<?xml version="1.0" encoding="utf-8"?>
<svg version="1.1" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" x="0px" y="0px"
viewBox="0 0 168.3 131.6" style="enable-background:new 0 0 168.3 131.6;" xml:space="preserve">
<style type="text/css">
.st0{fill:#FFFFFF;}
.st1{fill:#E22434;}
.st2{display:none;}
.st3{fill:#1B3663;}
.st4{fill:#00B1D1;}
.st5{fill:#BFE6EF;}
.st6{fill:#69CBE0;}
.st7{fill:#0080BD;}
.st8{fill:#005DAB;}
.st9{fill:#183660;}
.st10{fill:none;stroke:#183660;stroke-width:1.9441;stroke-miterlimit:10;}
.st11{fill:none;stroke:#183660;stroke-width:1.6201;stroke-miterlimit:10;}
.st12{fill:none;stroke:#183660;stroke-width:1.2961;stroke-miterlimit:10;}
.st13{fill:none;stroke:#C0E6EF;stroke-width:1.9441;stroke-miterlimit:10;}
.st14{fill:#C0E6EF;}
.st15{fill:none;stroke:#C0E6EF;stroke-width:1.2961;stroke-miterlimit:10;}
.st16{fill:none;stroke:#C0E6EF;stroke-width:1.6201;stroke-miterlimit:10;}
.st17{fill:none;stroke:#FFFFFF;stroke-width:1.9441;stroke-miterlimit:10;}
.st18{fill:none;stroke:#FFFFFF;stroke-width:1.2961;stroke-miterlimit:10;}
.st19{fill:none;stroke:#FFFFFF;stroke-width:1.6201;stroke-miterlimit:10;}
</style>
<g id="Layer_2">
</g>
<g id="Layer_1">
<g>
<g>
<g>
<path class="st3" d="M4.7,105l0.1,1.8c1.1-1.4,2.6-2.1,4.4-2.1c1.9,0,3.2,0.9,4,2.6c1.1-1.7,2.6-2.6,4.7-2.6
c3.3,0,5,2.3,5.1,6.9V124h-5v-12.1c0-1.1-0.2-1.9-0.5-2.4c-0.3-0.5-0.8-0.7-1.5-0.7c-0.9,0-1.6,0.6-2.1,1.7l0,0.6V124H9v-12.1
c0-1.1-0.1-1.9-0.4-2.4c-0.3-0.5-0.8-0.7-1.6-0.7c-0.9,0-1.5,0.5-2,1.4V124H0v-19H4.7z"/>
<path class="st3" d="M26.4,113.9c0-3.1,0.6-5.4,1.7-7s2.7-2.3,4.7-2.3c1.7,0,3.1,0.7,4,2L37,105h4.5v19c0,2.4-0.7,4.3-2,5.6
c-1.4,1.3-3.3,1.9-5.9,1.9c-1,0-2.1-0.2-3.3-0.6s-2-0.9-2.6-1.6l1.7-3.4c0.5,0.5,1.1,0.9,1.8,1.2c0.7,0.3,1.5,0.5,2.1,0.5
c1.1,0,1.9-0.3,2.4-0.8s0.7-1.4,0.7-2.6v-1.6c-0.9,1.2-2.2,1.9-3.7,1.9c-2,0-3.6-0.8-4.7-2.4s-1.7-3.8-1.7-6.7V113.9z
M31.4,115.2c0,1.8,0.2,3,0.7,3.8s1.2,1.2,2.2,1.2c1,0,1.8-0.4,2.3-1.1V110c-0.5-0.8-1.3-1.2-2.2-1.2c-1,0-1.7,0.4-2.2,1.2
s-0.7,2.1-0.7,3.9V115.2z"/>
<path class="st3" d="M50.1,105l0.1,1.8c1.1-1.4,2.6-2.1,4.4-2.1c1.9,0,3.2,0.9,4,2.6c1.1-1.7,2.6-2.6,4.7-2.6
c3.3,0,5,2.3,5.1,6.9V124h-5v-12.1c0-1.1-0.2-1.9-0.5-2.4s-0.8-0.7-1.5-0.7c-0.9,0-1.6,0.6-2.1,1.7l0,0.6V124h-5v-12.1
c0-1.1-0.1-1.9-0.4-2.4s-0.8-0.7-1.6-0.7c-0.9,0-1.5,0.5-2,1.4V124h-5v-19H50.1z"/>
<path class="st3" d="M78.2,100.3v4.7h2.5v3.7h-2.5v9.5c0,0.8,0.1,1.3,0.3,1.5c0.2,0.3,0.6,0.4,1.2,0.4c0.5,0,0.9,0,1.2-0.1
l0,3.9c-0.8,0.3-1.8,0.5-2.7,0.5c-3.2,0-4.8-1.8-4.9-5.5v-10.1h-2.2V105h2.2v-4.7H78.2z"/>
<path class="st4" d="M90.6,122.6c1.4,0,2.4-0.4,3.1-1.1c0.7-0.8,1.1-1.9,1.2-3.3h1.9c-0.1,1.9-0.7,3.5-1.9,4.6
c-1.1,1.1-2.6,1.7-4.3,1.7c-2.3,0-4-0.7-5.1-2.2s-1.7-3.6-1.8-6.4v-2.3c0-2.9,0.6-5.1,1.7-6.6s2.9-2.2,5.1-2.2
c1.9,0,3.4,0.6,4.5,1.8s1.7,2.9,1.7,5H95c-0.1-1.6-0.5-2.8-1.2-3.6s-1.8-1.3-3.1-1.3c-1.7,0-2.9,0.6-3.7,1.7s-1.2,2.9-1.2,5.2
v2.2c0,2.4,0.4,4.2,1.2,5.3S89,122.6,90.6,122.6z"/>
<path class="st4" d="M100.5,113.6c0-2.7,0.6-4.9,1.9-6.5s2.9-2.4,5.1-2.4c2.2,0,3.9,0.8,5.1,2.4s1.9,3.7,1.9,6.5v2
c0,2.8-0.6,5-1.9,6.5s-2.9,2.3-5.1,2.3c-2.1,0-3.8-0.8-5-2.3s-1.9-3.6-1.9-6.3V113.6z M102.5,115.5c0,2.2,0.4,3.9,1.3,5.2
c0.9,1.3,2.1,1.9,3.7,1.9c1.6,0,2.8-0.6,3.7-1.8s1.3-2.9,1.3-5.2v-2c0-2.2-0.4-3.9-1.3-5.2c-0.9-1.3-2.1-1.9-3.7-1.9
c-1.5,0-2.7,0.6-3.6,1.8s-1.3,2.9-1.4,5.1V115.5z"/>
<path class="st4" d="M121.1,105l0.1,3c0.6-1,1.3-1.9,2.2-2.5s1.9-0.9,3-0.9c3.3,0,5,2.2,5.1,6.6V124h-1.9v-12.5
c0-1.7-0.3-3-0.9-3.8c-0.6-0.8-1.5-1.2-2.8-1.2c-1,0-1.9,0.4-2.8,1.2s-1.4,1.8-1.9,3.2V124h-2v-19H121.1z"/>
<path class="st4" d="M138.2,124v-17.3h-2.6V105h2.6v-2.5c0-1.9,0.5-3.3,1.3-4.3s2-1.5,3.5-1.5c0.7,0,1.3,0.1,1.9,0.3l-0.1,1.8
c-0.5-0.1-1-0.2-1.6-0.2c-0.9,0-1.7,0.4-2.2,1.1s-0.8,1.7-0.8,3v2.3h3.7v1.8h-3.7V124H138.2z"/>
<path class="st4" d="M148,99.5c0-0.4,0.1-0.7,0.3-1s0.5-0.4,0.9-0.4s0.7,0.1,1,0.4s0.3,0.6,0.3,1s-0.1,0.7-0.3,1s-0.5,0.4-1,0.4
s-0.7-0.1-0.9-0.4S148,99.9,148,99.5z M150.2,124h-2v-19h2V124z"/>
<path class="st4" d="M155.3,113.6c0-3,0.5-5.2,1.5-6.7s2.6-2.2,4.6-2.2c2.2,0,3.8,1,4.9,3l0.1-2.7h1.8v19.6c0,2.3-0.6,4-1.6,5.2
c-1.1,1.2-2.7,1.8-4.7,1.8c-1,0-2.1-0.3-3.2-0.8s-1.9-1.1-2.4-1.9l0.9-1.4c1.3,1.5,2.8,2.2,4.5,2.2c1.6,0,2.7-0.4,3.5-1.3
c0.7-0.8,1.1-2.1,1.1-3.8v-3c-1.1,1.8-2.7,2.7-4.9,2.7c-2,0-3.5-0.7-4.5-2.2s-1.6-3.6-1.6-6.5V113.6z M157.2,115.4
c0,2.4,0.4,4.2,1.1,5.4s1.9,1.7,3.5,1.7c2.1,0,3.6-1,4.5-3v-9.7c-0.9-2.2-2.4-3.3-4.4-3.3c-1.6,0-2.8,0.6-3.5,1.7
s-1.1,2.9-1.1,5.3V115.4z"/>
</g>
</g>
<g>
<g>
<g>
<line class="st11" x1="58.9" y1="20.1" x2="77.9" y2="35.6"/>
<g>
<polygon class="st9" points="57.6,23.2 55.3,17.1 61.7,18.2 "/>
</g>
</g>
</g>
<circle class="st5" cx="48" cy="10.1" r="10.1"/>
<g>
<g>
<line class="st11" x1="77.9" y1="35.1" x2="96.2" y2="66.7"/>
<g>
<polygon class="st9" points="93,67.5 98.6,70.7 98.6,64.2 "/>
</g>
</g>
</g>
<g>
<g>
<line class="st11" x1="92.3" y1="38.9" x2="114.4" y2="43.6"/>
<g>
<polygon class="st9" points="92.6,42.3 87.8,38 93.9,36 "/>
</g>
</g>
</g>
<g>
<g>
<line class="st11" x1="77.9" y1="35.6" x2="94.5" y2="25.1"/>
<g>
<polygon class="st9" points="95.4,28.3 98.4,22.6 91.9,22.9 "/>
</g>
</g>
</g>
<g>
<g>
<line class="st11" x1="48.5" y1="41.3" x2="63.5" y2="38.7"/>
<g>
<polygon class="st9" points="63.1,42.1 68.1,38 62,35.7 "/>
</g>
</g>
</g>
<g>
<g>
<line class="st11" x1="111.4" y1="66.3" x2="124.4" y2="45.2"/>
<g>
<polygon class="st9" points="114.6,67.2 109,70.2 109.1,63.8 "/>
</g>
</g>
</g>
<g>
<g>
<line class="st11" x1="115.6" y1="30.4" x2="124.4" y2="45.2"/>
<g>
<polygon class="st9" points="113.3,32.8 113.3,26.4 118.9,29.5 "/>
</g>
</g>
</g>
<circle class="st3" cx="77.9" cy="35.6" r="10.1"/>
<circle class="st4" cx="44.6" cy="42.4" r="10.1"/>
<circle class="st7" cx="107.4" cy="18.1" r="10.1"/>
<circle class="st8" cx="104.1" cy="79.1" r="10.1"/>
<circle class="st6" cx="124.4" cy="45.2" r="10.1"/>
</g>
</g>
</g>
</svg>

After

Width:  |  Height:  |  Size: 6.2 KiB

View File

@@ -0,0 +1,132 @@
<?xml version="1.0" encoding="utf-8"?>
<svg version="1.1" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" x="0px" y="0px"
viewBox="0 0 260.4 71.4" style="enable-background:new 0 0 260.4 71.4;" xml:space="preserve">
<style type="text/css">
.st0{fill:#FFFFFF;}
.st1{fill:#E22434;}
.st2{display:none;}
.st3{fill:#1B3663;}
.st4{fill:#00B1D1;}
.st5{fill:#BFE6EF;}
.st6{fill:#69CBE0;}
.st7{fill:#0080BD;}
.st8{fill:#005DAB;}
.st9{fill:#183660;}
.st10{fill:none;stroke:#183660;stroke-width:1.9441;stroke-miterlimit:10;}
.st11{fill:none;stroke:#183660;stroke-width:1.6201;stroke-miterlimit:10;}
.st12{fill:none;stroke:#183660;stroke-width:1.2961;stroke-miterlimit:10;}
.st13{fill:none;stroke:#C0E6EF;stroke-width:1.9441;stroke-miterlimit:10;}
.st14{fill:#C0E6EF;}
.st15{fill:none;stroke:#C0E6EF;stroke-width:1.2961;stroke-miterlimit:10;}
.st16{fill:none;stroke:#C0E6EF;stroke-width:1.6201;stroke-miterlimit:10;}
.st17{fill:none;stroke:#FFFFFF;stroke-width:1.9441;stroke-miterlimit:10;}
.st18{fill:none;stroke:#FFFFFF;stroke-width:1.2961;stroke-miterlimit:10;}
.st19{fill:none;stroke:#FFFFFF;stroke-width:1.6201;stroke-miterlimit:10;}
</style>
<g id="Layer_2">
</g>
<g id="Layer_1">
<g>
<g>
<g>
<path class="st3" d="M96.7,25.7l0.1,1.8c1.1-1.4,2.6-2.1,4.4-2.1c1.9,0,3.2,0.9,4,2.6c1.1-1.7,2.6-2.6,4.7-2.6
c3.3,0,5,2.3,5.1,6.9v12.5h-5V32.6c0-1.1-0.2-1.9-0.5-2.4s-0.8-0.7-1.5-0.7c-0.9,0-1.6,0.6-2.1,1.7l0,0.6v12.9h-5V32.6
c0-1.1-0.1-1.9-0.4-2.4s-0.8-0.7-1.6-0.7c-0.9,0-1.5,0.5-2,1.4v13.8h-5v-19H96.7z"/>
<path class="st3" d="M118.5,34.6c0-3.1,0.6-5.4,1.7-7s2.7-2.3,4.7-2.3c1.7,0,3.1,0.7,4,2l0.2-1.7h4.5v19c0,2.4-0.7,4.3-2,5.6
c-1.4,1.3-3.3,1.9-5.9,1.9c-1,0-2.1-0.2-3.3-0.6s-2-0.9-2.6-1.6l1.7-3.4c0.5,0.5,1.1,0.9,1.8,1.2c0.7,0.3,1.5,0.5,2.1,0.5
c1.1,0,1.9-0.3,2.4-0.8s0.7-1.4,0.7-2.6v-1.6c-0.9,1.2-2.2,1.9-3.7,1.9c-2,0-3.6-0.8-4.7-2.4s-1.7-3.8-1.7-6.7V34.6z
M123.5,35.9c0,1.8,0.2,3,0.7,3.8s1.2,1.2,2.2,1.2c1,0,1.8-0.4,2.3-1.1v-9.1c-0.5-0.8-1.3-1.2-2.2-1.2c-1,0-1.7,0.4-2.2,1.2
s-0.7,2.1-0.7,3.9V35.9z"/>
<path class="st3" d="M142.2,25.7l0.1,1.8c1.1-1.4,2.6-2.1,4.4-2.1c1.9,0,3.2,0.9,4,2.6c1.1-1.7,2.6-2.6,4.7-2.6
c3.3,0,5,2.3,5.1,6.9v12.5h-5V32.6c0-1.1-0.2-1.9-0.5-2.4c-0.3-0.5-0.8-0.7-1.5-0.7c-0.9,0-1.6,0.6-2.1,1.7l0,0.6v12.9h-5V32.6
c0-1.1-0.1-1.9-0.4-2.4c-0.3-0.5-0.8-0.7-1.6-0.7c-0.9,0-1.5,0.5-2,1.4v13.8h-5v-19H142.2z"/>
<path class="st3" d="M170.3,21v4.7h2.5v3.7h-2.5v9.5c0,0.8,0.1,1.3,0.3,1.5c0.2,0.3,0.6,0.4,1.2,0.4c0.5,0,0.9,0,1.2-0.1l0,3.9
c-0.8,0.3-1.8,0.5-2.7,0.5c-3.2,0-4.8-1.8-4.9-5.5V29.4h-2.2v-3.7h2.2V21H170.3z"/>
<path class="st4" d="M182.7,43.2c1.4,0,2.4-0.4,3.1-1.1s1.1-1.8,1.2-3.3h1.9c-0.1,1.9-0.7,3.5-1.9,4.6s-2.6,1.7-4.3,1.7
c-2.3,0-4-0.7-5.1-2.2s-1.7-3.6-1.8-6.4v-2.3c0-2.9,0.6-5.1,1.7-6.6s2.9-2.2,5.1-2.2c1.9,0,3.4,0.6,4.5,1.8s1.7,2.9,1.7,5H187
c-0.1-1.6-0.5-2.8-1.2-3.6s-1.8-1.3-3.1-1.3c-1.7,0-2.9,0.6-3.7,1.7c-0.8,1.1-1.2,2.9-1.2,5.2v2.2c0,2.4,0.4,4.2,1.2,5.3
S181,43.2,182.7,43.2z"/>
<path class="st4" d="M192.6,34.2c0-2.7,0.6-4.9,1.9-6.5s2.9-2.4,5.1-2.4c2.2,0,3.9,0.8,5.1,2.4c1.2,1.6,1.9,3.7,1.9,6.5v2
c0,2.8-0.6,5-1.9,6.5s-2.9,2.3-5.1,2.3s-3.8-0.8-5-2.3s-1.9-3.6-1.9-6.3V34.2z M194.6,36.2c0,2.2,0.4,3.9,1.3,5.2
c0.9,1.3,2.1,1.9,3.7,1.9c1.6,0,2.8-0.6,3.7-1.8c0.8-1.2,1.3-2.9,1.3-5.2v-2c0-2.2-0.4-3.9-1.3-5.2c-0.9-1.3-2.1-1.9-3.7-1.9
c-1.5,0-2.7,0.6-3.6,1.8c-0.9,1.2-1.3,2.9-1.4,5.1V36.2z"/>
<path class="st4" d="M213.2,25.7l0.1,3c0.6-1,1.3-1.9,2.2-2.5s1.9-0.9,3-0.9c3.3,0,5,2.2,5.1,6.6v12.7h-1.9V32.2
c0-1.7-0.3-3-0.9-3.8c-0.6-0.8-1.5-1.2-2.8-1.2c-1,0-1.9,0.4-2.8,1.2s-1.4,1.8-1.9,3.2v13.2h-2v-19H213.2z"/>
<path class="st4" d="M230.3,44.7V27.4h-2.6v-1.8h2.6v-2.5c0-1.9,0.5-3.3,1.3-4.3s2-1.5,3.5-1.5c0.7,0,1.3,0.1,1.9,0.3l-0.1,1.8
c-0.5-0.1-1-0.2-1.6-0.2c-0.9,0-1.7,0.4-2.2,1.1s-0.8,1.7-0.8,3v2.3h3.7v1.8h-3.7v17.3H230.3z"/>
<path class="st4" d="M240.1,20.2c0-0.4,0.1-0.7,0.3-1s0.5-0.4,0.9-0.4s0.7,0.1,1,0.4s0.3,0.6,0.3,1s-0.1,0.7-0.3,1
s-0.5,0.4-1,0.4s-0.7-0.1-0.9-0.4S240.1,20.6,240.1,20.2z M242.3,44.7h-2v-19h2V44.7z"/>
<path class="st4" d="M247.4,34.3c0-3,0.5-5.2,1.5-6.7s2.6-2.2,4.6-2.2c2.2,0,3.8,1,4.9,3l0.1-2.7h1.8v19.6c0,2.3-0.6,4-1.6,5.2
s-2.7,1.8-4.7,1.8c-1,0-2.1-0.3-3.2-0.8s-1.9-1.1-2.4-1.9l0.9-1.4c1.3,1.5,2.8,2.2,4.5,2.2c1.6,0,2.7-0.4,3.5-1.3
c0.7-0.8,1.1-2.1,1.1-3.8v-3c-1.1,1.8-2.7,2.7-4.9,2.7c-2,0-3.5-0.7-4.5-2.2s-1.6-3.6-1.6-6.5V34.3z M249.3,36.1
c0,2.4,0.4,4.2,1.1,5.4s1.9,1.7,3.5,1.7c2.1,0,3.6-1,4.5-3v-9.7c-0.9-2.2-2.4-3.3-4.4-3.3c-1.6,0-2.8,0.6-3.5,1.7
s-1.1,2.9-1.1,5.3V36.1z"/>
</g>
</g>
<g>
<g>
<g>
<line class="st12" x1="19.5" y1="16" x2="34.7" y2="28.5"/>
<g>
<polygon class="st9" points="18.4,18.5 16.6,13.7 21.7,14.5 "/>
</g>
</g>
</g>
<circle class="st5" cx="10.8" cy="8.1" r="8.1"/>
<g>
<g>
<line class="st12" x1="34.7" y1="28" x2="49.4" y2="53.3"/>
<g>
<polygon class="st9" points="46.8,54 51.2,56.5 51.2,51.4 "/>
</g>
</g>
</g>
<g>
<g>
<line class="st12" x1="46.2" y1="31.1" x2="63.9" y2="34.9"/>
<g>
<polygon class="st9" points="46.4,33.8 42.6,30.4 47.5,28.8 "/>
</g>
</g>
</g>
<g>
<g>
<line class="st12" x1="34.7" y1="28.5" x2="47.9" y2="20.1"/>
<g>
<polygon class="st9" points="48.7,22.7 51.1,18.1 45.9,18.3 "/>
</g>
</g>
</g>
<g>
<g>
<line class="st12" x1="11.2" y1="33.1" x2="23.2" y2="31"/>
<g>
<polygon class="st9" points="22.9,33.7 26.8,30.4 22,28.6 "/>
</g>
</g>
</g>
<g>
<g>
<line class="st12" x1="61.5" y1="53" x2="71.9" y2="36.1"/>
<g>
<polygon class="st9" points="64.1,53.7 59.5,56.2 59.7,51 "/>
</g>
</g>
</g>
<g>
<g>
<line class="st12" x1="64.2" y1="24.2" x2="70.7" y2="33.9"/>
<g>
<polygon class="st9" points="62.5,26.3 62.2,21.1 66.8,23.4 "/>
</g>
</g>
</g>
<circle class="st3" cx="34.7" cy="28.5" r="8.1"/>
<circle class="st4" cx="8.1" cy="33.9" r="8.1"/>
<circle class="st7" cx="58.3" cy="14.5" r="8.1"/>
<circle class="st8" cx="55.7" cy="63.3" r="8.1"/>
<circle class="st6" cx="71.9" cy="36.1" r="8.1"/>
</g>
</g>
</g>
</svg>

After

Width:  |  Height:  |  Size: 6.2 KiB

View File

@@ -0,0 +1,94 @@
<?xml version="1.0" encoding="utf-8"?>
<svg version="1.1" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" x="0px" y="0px"
viewBox="0 0 120 107.1" style="enable-background:new 0 0 120 107.1;" xml:space="preserve">
<style type="text/css">
.st0{fill:#FFFFFF;}
.st1{fill:#E22434;}
.st2{display:none;}
.st3{fill:#1B3663;}
.st4{fill:#00B1D1;}
.st5{fill:#BFE6EF;}
.st6{fill:#69CBE0;}
.st7{fill:#0080BD;}
.st8{fill:#005DAB;}
.st9{fill:#183660;}
.st10{fill:none;stroke:#183660;stroke-width:1.9441;stroke-miterlimit:10;}
.st11{fill:none;stroke:#183660;stroke-width:1.6201;stroke-miterlimit:10;}
.st12{fill:none;stroke:#183660;stroke-width:1.2961;stroke-miterlimit:10;}
.st13{fill:none;stroke:#C0E6EF;stroke-width:1.9441;stroke-miterlimit:10;}
.st14{fill:#C0E6EF;}
.st15{fill:none;stroke:#C0E6EF;stroke-width:1.2961;stroke-miterlimit:10;}
.st16{fill:none;stroke:#C0E6EF;stroke-width:1.6201;stroke-miterlimit:10;}
.st17{fill:none;stroke:#FFFFFF;stroke-width:1.9441;stroke-miterlimit:10;}
.st18{fill:none;stroke:#FFFFFF;stroke-width:1.2961;stroke-miterlimit:10;}
.st19{fill:none;stroke:#FFFFFF;stroke-width:1.6201;stroke-miterlimit:10;}
</style>
<g id="Layer_2">
</g>
<g id="Layer_1">
<g>
<g>
<g>
<line class="st13" x1="29.2" y1="24.1" x2="52.1" y2="42.8"/>
<g>
<polygon class="st14" points="27.7,27.8 24.9,20.5 32.6,21.8 "/>
</g>
</g>
</g>
<circle class="st3" cx="16.1" cy="12.2" r="12.1"/>
<g>
<g>
<line class="st13" x1="52.1" y1="42.1" x2="74.1" y2="80"/>
<g>
<polygon class="st14" points="70.1,80.9 76.9,84.8 76.8,77.1 "/>
</g>
</g>
</g>
<g>
<g>
<line class="st13" x1="69.4" y1="46.7" x2="95.8" y2="52.4"/>
<g>
<polygon class="st14" points="69.7,50.7 63.9,45.6 71.3,43.2 "/>
</g>
</g>
</g>
<g>
<g>
<line class="st13" x1="52.1" y1="42.8" x2="71.9" y2="30.1"/>
<g>
<polygon class="st14" points="73.1,34 76.6,27.1 68.9,27.5 "/>
</g>
</g>
</g>
<g>
<g>
<line class="st13" x1="16.8" y1="49.6" x2="34.8" y2="46.5"/>
<g>
<polygon class="st14" points="34.3,50.5 40.3,45.6 33,42.9 "/>
</g>
</g>
</g>
<g>
<g>
<line class="st13" x1="92.3" y1="79.5" x2="107.8" y2="54.2"/>
<g>
<polygon class="st14" points="96.1,80.6 89.3,84.3 89.5,76.5 "/>
</g>
</g>
</g>
<g>
<g>
<line class="st13" x1="97.2" y1="36.3" x2="107.8" y2="54.2"/>
<g>
<polygon class="st14" points="94.4,39.3 94.3,31.5 101.1,35.3 "/>
</g>
</g>
</g>
<circle class="st5" cx="52.1" cy="42.8" r="12.1"/>
<circle class="st4" cx="12.2" cy="50.8" r="12.1"/>
<circle class="st7" cx="87.5" cy="21.7" r="12.1"/>
<circle class="st8" cx="83.5" cy="95" r="12.1"/>
<circle class="st6" cx="107.8" cy="54.2" r="12.1"/>
</g>
</g>
</svg>

After

Width:  |  Height:  |  Size: 2.8 KiB

View File

@@ -0,0 +1,132 @@
<?xml version="1.0" encoding="utf-8"?>
<svg version="1.1" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" x="0px" y="0px"
viewBox="0 0 168.3 133" style="enable-background:new 0 0 168.3 133;" xml:space="preserve">
<style type="text/css">
.st0{fill:#FFFFFF;}
.st1{fill:#E22434;}
.st2{display:none;}
.st3{fill:#1B3663;}
.st4{fill:#00B1D1;}
.st5{fill:#BFE6EF;}
.st6{fill:#69CBE0;}
.st7{fill:#0080BD;}
.st8{fill:#005DAB;}
.st9{fill:#183660;}
.st10{fill:none;stroke:#183660;stroke-width:1.9441;stroke-miterlimit:10;}
.st11{fill:none;stroke:#183660;stroke-width:1.6201;stroke-miterlimit:10;}
.st12{fill:none;stroke:#183660;stroke-width:1.2961;stroke-miterlimit:10;}
.st13{fill:none;stroke:#C0E6EF;stroke-width:1.9441;stroke-miterlimit:10;}
.st14{fill:#C0E6EF;}
.st15{fill:none;stroke:#C0E6EF;stroke-width:1.2961;stroke-miterlimit:10;}
.st16{fill:none;stroke:#C0E6EF;stroke-width:1.6201;stroke-miterlimit:10;}
.st17{fill:none;stroke:#FFFFFF;stroke-width:1.9441;stroke-miterlimit:10;}
.st18{fill:none;stroke:#FFFFFF;stroke-width:1.2961;stroke-miterlimit:10;}
.st19{fill:none;stroke:#FFFFFF;stroke-width:1.6201;stroke-miterlimit:10;}
</style>
<g id="Layer_2">
</g>
<g id="Layer_1">
<g>
<g>
<g>
<path class="st0" d="M4.7,106.4l0.1,1.8c1.1-1.4,2.6-2.1,4.4-2.1c1.9,0,3.2,0.9,4,2.6c1.1-1.7,2.6-2.6,4.7-2.6
c3.3,0,5,2.3,5.1,6.9v12.5h-5v-12.1c0-1.1-0.2-1.9-0.5-2.4c-0.3-0.5-0.8-0.7-1.5-0.7c-0.9,0-1.6,0.6-2.1,1.7l0,0.6v12.9H9v-12.1
c0-1.1-0.1-1.9-0.4-2.4c-0.3-0.5-0.8-0.7-1.6-0.7c-0.9,0-1.5,0.5-2,1.4v13.8H0v-19H4.7z"/>
<path class="st0" d="M26.4,115.3c0-3.1,0.6-5.4,1.7-7s2.7-2.3,4.7-2.3c1.7,0,3.1,0.7,4,2l0.2-1.7h4.5v19c0,2.4-0.7,4.3-2,5.6
c-1.4,1.3-3.3,1.9-5.9,1.9c-1,0-2.1-0.2-3.3-0.6s-2-0.9-2.6-1.6l1.7-3.4c0.5,0.5,1.1,0.9,1.8,1.2c0.7,0.3,1.5,0.5,2.1,0.5
c1.1,0,1.9-0.3,2.4-0.8c0.5-0.5,0.7-1.4,0.7-2.6v-1.6c-0.9,1.2-2.2,1.9-3.7,1.9c-2,0-3.6-0.8-4.7-2.4s-1.7-3.8-1.7-6.7V115.3z
M31.4,116.6c0,1.8,0.2,3,0.7,3.8s1.2,1.2,2.2,1.2c1,0,1.8-0.4,2.3-1.1v-9.1c-0.5-0.8-1.3-1.2-2.2-1.2c-1,0-1.7,0.4-2.2,1.2
s-0.7,2.1-0.7,3.9V116.6z"/>
<path class="st0" d="M50.1,106.4l0.1,1.8c1.1-1.4,2.6-2.1,4.4-2.1c1.9,0,3.2,0.9,4,2.6c1.1-1.7,2.6-2.6,4.7-2.6
c3.3,0,5,2.3,5.1,6.9v12.5h-5v-12.1c0-1.1-0.2-1.9-0.5-2.4s-0.8-0.7-1.5-0.7c-0.9,0-1.6,0.6-2.1,1.7l0,0.6v12.9h-5v-12.1
c0-1.1-0.1-1.9-0.4-2.4s-0.8-0.7-1.6-0.7c-0.9,0-1.5,0.5-2,1.4v13.8h-5v-19H50.1z"/>
<path class="st0" d="M78.2,101.7v4.7h2.5v3.7h-2.5v9.5c0,0.8,0.1,1.3,0.3,1.5c0.2,0.3,0.6,0.4,1.2,0.4c0.5,0,0.9,0,1.2-0.1
l0,3.9c-0.8,0.3-1.8,0.5-2.7,0.5c-3.2,0-4.8-1.8-4.9-5.5v-10.1h-2.2v-3.7h2.2v-4.7H78.2z"/>
<path class="st4" d="M90.6,124c1.4,0,2.4-0.4,3.1-1.1c0.7-0.8,1.1-1.9,1.2-3.3h1.9c-0.1,1.9-0.7,3.5-1.9,4.6
c-1.1,1.1-2.6,1.7-4.3,1.7c-2.3,0-4-0.7-5.1-2.2s-1.7-3.6-1.8-6.4v-2.3c0-2.9,0.6-5.1,1.7-6.6s2.9-2.2,5.1-2.2
c1.9,0,3.4,0.6,4.5,1.8s1.7,2.9,1.7,5H95c-0.1-1.6-0.5-2.8-1.2-3.6s-1.8-1.3-3.1-1.3c-1.7,0-2.9,0.6-3.7,1.7s-1.2,2.9-1.2,5.2
v2.2c0,2.4,0.4,4.2,1.2,5.3S89,124,90.6,124z"/>
<path class="st4" d="M100.5,115c0-2.7,0.6-4.9,1.9-6.5s2.9-2.4,5.1-2.4c2.2,0,3.9,0.8,5.1,2.4s1.9,3.7,1.9,6.5v2
c0,2.8-0.6,5-1.9,6.5s-2.9,2.3-5.1,2.3c-2.1,0-3.8-0.8-5-2.3s-1.9-3.6-1.9-6.3V115z M102.5,116.9c0,2.2,0.4,3.9,1.3,5.2
c0.9,1.3,2.1,1.9,3.7,1.9c1.6,0,2.8-0.6,3.7-1.8s1.3-2.9,1.3-5.2v-2c0-2.2-0.4-3.9-1.3-5.2c-0.9-1.3-2.1-1.9-3.7-1.9
c-1.5,0-2.7,0.6-3.6,1.8s-1.3,2.9-1.4,5.1V116.9z"/>
<path class="st4" d="M121.1,106.4l0.1,3c0.6-1,1.3-1.9,2.2-2.5s1.9-0.9,3-0.9c3.3,0,5,2.2,5.1,6.6v12.7h-1.9v-12.5
c0-1.7-0.3-3-0.9-3.8c-0.6-0.8-1.5-1.2-2.8-1.2c-1,0-1.9,0.4-2.8,1.2s-1.4,1.8-1.9,3.2v13.2h-2v-19H121.1z"/>
<path class="st4" d="M138.2,125.4v-17.3h-2.6v-1.8h2.6v-2.5c0-1.9,0.5-3.3,1.3-4.3s2-1.5,3.5-1.5c0.7,0,1.3,0.1,1.9,0.3
l-0.1,1.8c-0.5-0.1-1-0.2-1.6-0.2c-0.9,0-1.7,0.4-2.2,1.1s-0.8,1.7-0.8,3v2.3h3.7v1.8h-3.7v17.3H138.2z"/>
<path class="st4" d="M148,100.9c0-0.4,0.1-0.7,0.3-1s0.5-0.4,0.9-0.4s0.7,0.1,1,0.4s0.3,0.6,0.3,1s-0.1,0.7-0.3,1
s-0.5,0.4-1,0.4s-0.7-0.1-0.9-0.4S148,101.3,148,100.9z M150.2,125.4h-2v-19h2V125.4z"/>
<path class="st4" d="M155.3,115c0-3,0.5-5.2,1.5-6.7s2.6-2.2,4.6-2.2c2.2,0,3.8,1,4.9,3l0.1-2.7h1.8V126c0,2.3-0.6,4-1.6,5.2
c-1.1,1.2-2.7,1.8-4.7,1.8c-1,0-2.1-0.3-3.2-0.8s-1.9-1.1-2.4-1.9l0.9-1.4c1.3,1.5,2.8,2.2,4.5,2.2c1.6,0,2.7-0.4,3.5-1.3
c0.7-0.8,1.1-2.1,1.1-3.8v-3c-1.1,1.8-2.7,2.7-4.9,2.7c-2,0-3.5-0.7-4.5-2.2s-1.6-3.6-1.6-6.5V115z M157.2,116.8
c0,2.4,0.4,4.2,1.1,5.4s1.9,1.7,3.5,1.7c2.1,0,3.6-1,4.5-3v-9.7c-0.9-2.2-2.4-3.3-4.4-3.3c-1.6,0-2.8,0.6-3.5,1.7
s-1.1,2.9-1.1,5.3V116.8z"/>
</g>
</g>
<g>
<g>
<g>
<line class="st16" x1="58.9" y1="20.1" x2="77.9" y2="35.6"/>
<g>
<polygon class="st14" points="57.6,23.2 55.3,17.1 61.7,18.2 "/>
</g>
</g>
</g>
<circle class="st3" cx="48" cy="10.1" r="10.1"/>
<g>
<g>
<line class="st16" x1="77.9" y1="35.1" x2="96.2" y2="66.7"/>
<g>
<polygon class="st14" points="93,67.5 98.6,70.7 98.6,64.2 "/>
</g>
</g>
</g>
<g>
<g>
<line class="st16" x1="92.3" y1="38.9" x2="114.4" y2="43.6"/>
<g>
<polygon class="st14" points="92.6,42.3 87.8,38 93.9,36 "/>
</g>
</g>
</g>
<g>
<g>
<line class="st16" x1="77.9" y1="35.6" x2="94.5" y2="25.1"/>
<g>
<polygon class="st14" points="95.4,28.3 98.4,22.6 91.9,22.9 "/>
</g>
</g>
</g>
<g>
<g>
<line class="st16" x1="48.5" y1="41.3" x2="63.5" y2="38.7"/>
<g>
<polygon class="st14" points="63.1,42.1 68.1,38 62,35.7 "/>
</g>
</g>
</g>
<g>
<g>
<line class="st16" x1="111.4" y1="66.3" x2="124.4" y2="45.2"/>
<g>
<polygon class="st14" points="114.6,67.2 109,70.2 109.1,63.8 "/>
</g>
</g>
</g>
<g>
<g>
<line class="st16" x1="115.5" y1="30.5" x2="124.4" y2="45.2"/>
<g>
<polygon class="st14" points="113.2,32.9 113.1,26.5 118.8,29.6 "/>
</g>
</g>
</g>
<circle class="st5" cx="77.9" cy="35.6" r="10.1"/>
<circle class="st4" cx="44.6" cy="42.4" r="10.1"/>
<circle class="st7" cx="107.4" cy="18.1" r="10.1"/>
<circle class="st8" cx="104.1" cy="79.1" r="10.1"/>
<circle class="st6" cx="124.4" cy="45.2" r="10.1"/>
</g>
</g>
</g>
</svg>

After

Width:  |  Height:  |  Size: 6.2 KiB

View File

@@ -0,0 +1,132 @@
<?xml version="1.0" encoding="utf-8"?>
<svg version="1.1" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" x="0px" y="0px"
viewBox="0 0 260.4 71.4" style="enable-background:new 0 0 260.4 71.4;" xml:space="preserve">
<style type="text/css">
.st0{fill:#FFFFFF;}
.st1{fill:#E22434;}
.st2{display:none;}
.st3{fill:#1B3663;}
.st4{fill:#00B1D1;}
.st5{fill:#BFE6EF;}
.st6{fill:#69CBE0;}
.st7{fill:#0080BD;}
.st8{fill:#005DAB;}
.st9{fill:#183660;}
.st10{fill:none;stroke:#183660;stroke-width:1.9441;stroke-miterlimit:10;}
.st11{fill:none;stroke:#183660;stroke-width:1.6201;stroke-miterlimit:10;}
.st12{fill:none;stroke:#183660;stroke-width:1.2961;stroke-miterlimit:10;}
.st13{fill:none;stroke:#C0E6EF;stroke-width:1.9441;stroke-miterlimit:10;}
.st14{fill:#C0E6EF;}
.st15{fill:none;stroke:#C0E6EF;stroke-width:1.2961;stroke-miterlimit:10;}
.st16{fill:none;stroke:#C0E6EF;stroke-width:1.6201;stroke-miterlimit:10;}
.st17{fill:none;stroke:#FFFFFF;stroke-width:1.9441;stroke-miterlimit:10;}
.st18{fill:none;stroke:#FFFFFF;stroke-width:1.2961;stroke-miterlimit:10;}
.st19{fill:none;stroke:#FFFFFF;stroke-width:1.6201;stroke-miterlimit:10;}
</style>
<g id="Layer_2">
</g>
<g id="Layer_1">
<g>
<g>
<g>
<path class="st0" d="M96.7,27.6l0.1,1.8c1.1-1.4,2.6-2.1,4.4-2.1c1.9,0,3.2,0.9,4,2.6c1.1-1.7,2.6-2.6,4.7-2.6
c3.3,0,5,2.3,5.1,6.9v12.5h-5V34.4c0-1.1-0.2-1.9-0.5-2.4c-0.3-0.5-0.8-0.7-1.5-0.7c-0.9,0-1.6,0.6-2.1,1.7l0,0.6v12.9h-5V34.5
c0-1.1-0.1-1.9-0.4-2.4c-0.3-0.5-0.8-0.7-1.6-0.7c-0.9,0-1.5,0.5-2,1.4v13.8h-5v-19H96.7z"/>
<path class="st0" d="M118.5,36.5c0-3.1,0.6-5.4,1.7-7s2.7-2.3,4.7-2.3c1.7,0,3.1,0.7,4,2l0.2-1.7h4.5v19c0,2.4-0.7,4.3-2,5.6
c-1.4,1.3-3.3,1.9-5.9,1.9c-1,0-2.1-0.2-3.3-0.6s-2-0.9-2.6-1.6l1.7-3.4c0.5,0.5,1.1,0.9,1.8,1.2c0.7,0.3,1.5,0.5,2.1,0.5
c1.1,0,1.9-0.3,2.4-0.8s0.7-1.4,0.7-2.6v-1.6c-0.9,1.2-2.2,1.9-3.7,1.9c-2,0-3.6-0.8-4.7-2.4s-1.7-3.8-1.7-6.7V36.5z
M123.5,37.7c0,1.8,0.2,3,0.7,3.8s1.2,1.2,2.2,1.2c1,0,1.8-0.4,2.3-1.1v-9.1c-0.5-0.8-1.3-1.2-2.2-1.2c-1,0-1.7,0.4-2.2,1.2
c-0.5,0.8-0.7,2.1-0.7,3.9V37.7z"/>
<path class="st0" d="M142.2,27.6l0.1,1.8c1.1-1.4,2.6-2.1,4.4-2.1c1.9,0,3.2,0.9,4,2.6c1.1-1.7,2.6-2.6,4.7-2.6
c3.3,0,5,2.3,5.1,6.9v12.5h-5V34.4c0-1.1-0.2-1.9-0.5-2.4c-0.3-0.5-0.8-0.7-1.5-0.7c-0.9,0-1.6,0.6-2.1,1.7l0,0.6v12.9h-5V34.5
c0-1.1-0.1-1.9-0.4-2.4c-0.3-0.5-0.8-0.7-1.6-0.7c-0.9,0-1.5,0.5-2,1.4v13.8h-5v-19H142.2z"/>
<path class="st0" d="M170.3,22.9v4.7h2.5v3.7h-2.5v9.5c0,0.8,0.1,1.3,0.3,1.5c0.2,0.3,0.6,0.4,1.2,0.4c0.5,0,0.9,0,1.2-0.1
l0,3.9c-0.8,0.3-1.8,0.5-2.7,0.5c-3.2,0-4.8-1.8-4.9-5.5V31.3h-2.2v-3.7h2.2v-4.7H170.3z"/>
<path class="st4" d="M182.7,45.1c1.4,0,2.4-0.4,3.1-1.1c0.7-0.8,1.1-1.9,1.2-3.3h1.9c-0.1,1.9-0.7,3.5-1.9,4.6s-2.6,1.7-4.3,1.7
c-2.3,0-4-0.7-5.1-2.2s-1.7-3.6-1.8-6.4V36c0-2.9,0.6-5.1,1.7-6.6s2.9-2.2,5.1-2.2c1.9,0,3.4,0.6,4.5,1.8s1.7,2.9,1.7,5H187
c-0.1-1.6-0.5-2.8-1.2-3.6s-1.8-1.3-3.1-1.3c-1.7,0-2.9,0.6-3.7,1.7s-1.2,2.9-1.2,5.2v2.2c0,2.4,0.4,4.2,1.2,5.3
S181,45.1,182.7,45.1z"/>
<path class="st4" d="M192.6,36.1c0-2.7,0.6-4.9,1.9-6.5s2.9-2.4,5.1-2.4c2.2,0,3.9,0.8,5.1,2.4c1.2,1.6,1.9,3.7,1.9,6.5v2
c0,2.8-0.6,5-1.9,6.5s-2.9,2.3-5.1,2.3s-3.8-0.8-5-2.3s-1.9-3.6-1.9-6.3V36.1z M194.6,38.1c0,2.2,0.4,3.9,1.3,5.2
c0.9,1.3,2.1,1.9,3.7,1.9c1.6,0,2.8-0.6,3.7-1.8c0.8-1.2,1.3-2.9,1.3-5.2v-2c0-2.2-0.4-3.9-1.3-5.2c-0.9-1.3-2.1-1.9-3.7-1.9
c-1.5,0-2.7,0.6-3.6,1.8c-0.9,1.2-1.3,2.9-1.4,5.1V38.1z"/>
<path class="st4" d="M213.2,27.6l0.1,3c0.6-1,1.3-1.9,2.2-2.5s1.9-0.9,3-0.9c3.3,0,5,2.2,5.1,6.6v12.7h-1.9V34.1
c0-1.7-0.3-3-0.9-3.8c-0.6-0.8-1.5-1.2-2.8-1.2c-1,0-1.9,0.4-2.8,1.2s-1.4,1.8-1.9,3.2v13.2h-2v-19H213.2z"/>
<path class="st4" d="M230.3,46.6V29.3h-2.6v-1.8h2.6v-2.5c0-1.9,0.5-3.3,1.3-4.3s2-1.5,3.5-1.5c0.7,0,1.3,0.1,1.9,0.3l-0.1,1.8
c-0.5-0.1-1-0.2-1.6-0.2c-0.9,0-1.7,0.4-2.2,1.1s-0.8,1.7-0.8,3v2.3h3.7v1.8h-3.7v17.3H230.3z"/>
<path class="st4" d="M240.1,22.1c0-0.4,0.1-0.7,0.3-1s0.5-0.4,0.9-0.4s0.7,0.1,1,0.4s0.3,0.6,0.3,1s-0.1,0.7-0.3,1
s-0.5,0.4-1,0.4s-0.7-0.1-0.9-0.4S240.1,22.5,240.1,22.1z M242.3,46.6h-2v-19h2V46.6z"/>
<path class="st4" d="M247.4,36.2c0-3,0.5-5.2,1.5-6.7s2.6-2.2,4.6-2.2c2.2,0,3.8,1,4.9,3l0.1-2.7h1.8v19.6c0,2.3-0.6,4-1.6,5.2
s-2.7,1.8-4.7,1.8c-1,0-2.1-0.3-3.2-0.8s-1.9-1.1-2.4-1.9l0.9-1.4c1.3,1.5,2.8,2.2,4.5,2.2c1.6,0,2.7-0.4,3.5-1.3
c0.7-0.8,1.1-2.1,1.1-3.8v-3c-1.1,1.8-2.7,2.7-4.9,2.7c-2,0-3.5-0.7-4.5-2.2s-1.6-3.6-1.6-6.5V36.2z M249.3,38
c0,2.4,0.4,4.2,1.1,5.4s1.9,1.7,3.5,1.7c2.1,0,3.6-1,4.5-3v-9.7c-0.9-2.2-2.4-3.3-4.4-3.3c-1.6,0-2.8,0.6-3.5,1.7
s-1.1,2.9-1.1,5.3V38z"/>
</g>
</g>
<g>
<g>
<g>
<line class="st15" x1="19.5" y1="16" x2="34.7" y2="28.5"/>
<g>
<polygon class="st14" points="18.4,18.5 16.6,13.7 21.7,14.5 "/>
</g>
</g>
</g>
<circle class="st3" cx="10.8" cy="8.1" r="8.1"/>
<g>
<g>
<line class="st15" x1="34.7" y1="28" x2="49.4" y2="53.3"/>
<g>
<polygon class="st14" points="46.8,54 51.2,56.5 51.2,51.4 "/>
</g>
</g>
</g>
<g>
<g>
<line class="st15" x1="46.2" y1="31.1" x2="63.9" y2="34.9"/>
<g>
<polygon class="st14" points="46.4,33.8 42.6,30.4 47.5,28.8 "/>
</g>
</g>
</g>
<g>
<g>
<line class="st15" x1="34.7" y1="28.5" x2="47.9" y2="20.1"/>
<g>
<polygon class="st14" points="48.7,22.7 51.1,18.1 45.9,18.3 "/>
</g>
</g>
</g>
<g>
<g>
<line class="st15" x1="11.2" y1="33.1" x2="23.2" y2="31"/>
<g>
<polygon class="st14" points="22.9,33.7 26.8,30.4 22,28.6 "/>
</g>
</g>
</g>
<g>
<g>
<line class="st15" x1="61.5" y1="53" x2="71.9" y2="36.1"/>
<g>
<polygon class="st14" points="64.1,53.7 59.5,56.2 59.7,51 "/>
</g>
</g>
</g>
<g>
<g>
<line class="st15" x1="64.8" y1="24.4" x2="71.9" y2="36.1"/>
<g>
<polygon class="st14" points="63,26.4 62.9,21.2 67.4,23.7 "/>
</g>
</g>
</g>
<circle class="st5" cx="34.7" cy="28.5" r="8.1"/>
<circle class="st4" cx="8.1" cy="33.9" r="8.1"/>
<circle class="st7" cx="58.3" cy="14.5" r="8.1"/>
<circle class="st8" cx="55.7" cy="63.3" r="8.1"/>
<circle class="st6" cx="71.9" cy="36.1" r="8.1"/>
</g>
</g>
</g>
</svg>

After

Width:  |  Height:  |  Size: 6.2 KiB

View File

@@ -0,0 +1,94 @@
<?xml version="1.0" encoding="utf-8"?>
<svg version="1.1" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" x="0px" y="0px"
viewBox="0 0 120 107.1" style="enable-background:new 0 0 120 107.1;" xml:space="preserve">
<style type="text/css">
.st0{fill:#FFFFFF;}
.st1{fill:#E22434;}
.st2{display:none;}
.st3{fill:#1B3663;}
.st4{fill:#00B1D1;}
.st5{fill:#BFE6EF;}
.st6{fill:#69CBE0;}
.st7{fill:#0080BD;}
.st8{fill:#005DAB;}
.st9{fill:#183660;}
.st10{fill:none;stroke:#183660;stroke-width:1.9441;stroke-miterlimit:10;}
.st11{fill:none;stroke:#183660;stroke-width:1.6201;stroke-miterlimit:10;}
.st12{fill:none;stroke:#183660;stroke-width:1.2961;stroke-miterlimit:10;}
.st13{fill:none;stroke:#C0E6EF;stroke-width:1.9441;stroke-miterlimit:10;}
.st14{fill:#C0E6EF;}
.st15{fill:none;stroke:#C0E6EF;stroke-width:1.2961;stroke-miterlimit:10;}
.st16{fill:none;stroke:#C0E6EF;stroke-width:1.6201;stroke-miterlimit:10;}
.st17{fill:none;stroke:#FFFFFF;stroke-width:1.9441;stroke-miterlimit:10;}
.st18{fill:none;stroke:#FFFFFF;stroke-width:1.2961;stroke-miterlimit:10;}
.st19{fill:none;stroke:#FFFFFF;stroke-width:1.6201;stroke-miterlimit:10;}
</style>
<g id="Layer_2">
</g>
<g id="Layer_1">
<g>
<g>
<g>
<line class="st17" x1="29.2" y1="24.1" x2="52.1" y2="42.8"/>
<g>
<polygon class="st0" points="27.7,27.8 24.9,20.5 32.6,21.8 "/>
</g>
</g>
</g>
<circle class="st0" cx="16.1" cy="12.2" r="12.1"/>
<g>
<g>
<line class="st17" x1="52.1" y1="42.1" x2="74.1" y2="80"/>
<g>
<polygon class="st0" points="70.1,80.9 76.9,84.8 76.8,77.1 "/>
</g>
</g>
</g>
<g>
<g>
<line class="st17" x1="69.4" y1="46.7" x2="95.8" y2="52.4"/>
<g>
<polygon class="st0" points="69.7,50.7 63.9,45.6 71.3,43.2 "/>
</g>
</g>
</g>
<g>
<g>
<line class="st17" x1="52.1" y1="42.8" x2="71.9" y2="30.1"/>
<g>
<polygon class="st0" points="73.1,34 76.6,27.1 68.9,27.5 "/>
</g>
</g>
</g>
<g>
<g>
<line class="st17" x1="16.8" y1="49.6" x2="34.8" y2="46.5"/>
<g>
<polygon class="st0" points="34.3,50.5 40.3,45.6 33,42.9 "/>
</g>
</g>
</g>
<g>
<g>
<line class="st17" x1="92.3" y1="79.5" x2="107.8" y2="54.2"/>
<g>
<polygon class="st0" points="96.1,80.6 89.3,84.3 89.5,76.5 "/>
</g>
</g>
</g>
<g>
<g>
<line class="st17" x1="97.2" y1="36.6" x2="107.8" y2="54.2"/>
<g>
<polygon class="st0" points="94.4,39.5 94.3,31.8 101.1,35.5 "/>
</g>
</g>
</g>
<circle class="st0" cx="52.1" cy="42.8" r="12.1"/>
<circle class="st0" cx="12.2" cy="50.8" r="12.1"/>
<circle class="st0" cx="87.5" cy="21.7" r="12.1"/>
<circle class="st0" cx="83.5" cy="95" r="12.1"/>
<circle class="st0" cx="107.8" cy="54.2" r="12.1"/>
</g>
</g>
</svg>

After

Width:  |  Height:  |  Size: 2.8 KiB

View File

@@ -0,0 +1,132 @@
<?xml version="1.0" encoding="utf-8"?>
<svg version="1.1" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" x="0px" y="0px"
viewBox="0 0 168.3 131.4" style="enable-background:new 0 0 168.3 131.4;" xml:space="preserve">
<style type="text/css">
.st0{fill:#FFFFFF;}
.st1{fill:#E22434;}
.st2{display:none;}
.st3{fill:#1B3663;}
.st4{fill:#00B1D1;}
.st5{fill:#BFE6EF;}
.st6{fill:#69CBE0;}
.st7{fill:#0080BD;}
.st8{fill:#005DAB;}
.st9{fill:#183660;}
.st10{fill:none;stroke:#183660;stroke-width:1.9441;stroke-miterlimit:10;}
.st11{fill:none;stroke:#183660;stroke-width:1.6201;stroke-miterlimit:10;}
.st12{fill:none;stroke:#183660;stroke-width:1.2961;stroke-miterlimit:10;}
.st13{fill:none;stroke:#C0E6EF;stroke-width:1.9441;stroke-miterlimit:10;}
.st14{fill:#C0E6EF;}
.st15{fill:none;stroke:#C0E6EF;stroke-width:1.2961;stroke-miterlimit:10;}
.st16{fill:none;stroke:#C0E6EF;stroke-width:1.6201;stroke-miterlimit:10;}
.st17{fill:none;stroke:#FFFFFF;stroke-width:1.9441;stroke-miterlimit:10;}
.st18{fill:none;stroke:#FFFFFF;stroke-width:1.2961;stroke-miterlimit:10;}
.st19{fill:none;stroke:#FFFFFF;stroke-width:1.6201;stroke-miterlimit:10;}
</style>
<g id="Layer_2">
</g>
<g id="Layer_1">
<g>
<g>
<g>
<path class="st0" d="M4.7,104.8l0.1,1.8c1.1-1.4,2.6-2.1,4.4-2.1c1.9,0,3.2,0.9,4,2.6c1.1-1.7,2.6-2.6,4.7-2.6
c3.3,0,5,2.3,5.1,6.9v12.5h-5v-12.1c0-1.1-0.2-1.9-0.5-2.4c-0.3-0.5-0.8-0.7-1.5-0.7c-0.9,0-1.6,0.6-2.1,1.7l0,0.6v12.9H9v-12.1
c0-1.1-0.1-1.9-0.4-2.4c-0.3-0.5-0.8-0.7-1.6-0.7c-0.9,0-1.5,0.5-2,1.4v13.8H0v-19H4.7z"/>
<path class="st0" d="M26.4,113.8c0-3.1,0.6-5.4,1.7-7s2.7-2.3,4.7-2.3c1.7,0,3.1,0.7,4,2l0.2-1.7h4.5v19c0,2.4-0.7,4.3-2,5.6
c-1.4,1.3-3.3,1.9-5.9,1.9c-1,0-2.1-0.2-3.3-0.6s-2-0.9-2.6-1.6l1.7-3.4c0.5,0.5,1.1,0.9,1.8,1.2c0.7,0.3,1.5,0.5,2.1,0.5
c1.1,0,1.9-0.3,2.4-0.8s0.7-1.4,0.7-2.6v-1.6c-0.9,1.2-2.2,1.9-3.7,1.9c-2,0-3.6-0.8-4.7-2.4s-1.7-3.8-1.7-6.7V113.8z M31.4,115
c0,1.8,0.2,3,0.7,3.8s1.2,1.2,2.2,1.2c1,0,1.8-0.4,2.3-1.1v-9.1c-0.5-0.8-1.3-1.2-2.2-1.2c-1,0-1.7,0.4-2.2,1.2
s-0.7,2.1-0.7,3.9V115z"/>
<path class="st0" d="M50.1,104.8l0.1,1.8c1.1-1.4,2.6-2.1,4.4-2.1c1.9,0,3.2,0.9,4,2.6c1.1-1.7,2.6-2.6,4.7-2.6
c3.3,0,5,2.3,5.1,6.9v12.5h-5v-12.1c0-1.1-0.2-1.9-0.5-2.4s-0.8-0.7-1.5-0.7c-0.9,0-1.6,0.6-2.1,1.7l0,0.6v12.9h-5v-12.1
c0-1.1-0.1-1.9-0.4-2.4s-0.8-0.7-1.6-0.7c-0.9,0-1.5,0.5-2,1.4v13.8h-5v-19H50.1z"/>
<path class="st0" d="M78.2,100.2v4.7h2.5v3.7h-2.5v9.5c0,0.8,0.1,1.3,0.3,1.5c0.2,0.3,0.6,0.4,1.2,0.4c0.5,0,0.9,0,1.2-0.1
l0,3.9c-0.8,0.3-1.8,0.5-2.7,0.5c-3.2,0-4.8-1.8-4.9-5.5v-10.1h-2.2v-3.7h2.2v-4.7H78.2z"/>
<path class="st0" d="M90.6,122.4c1.4,0,2.4-0.4,3.1-1.1c0.7-0.8,1.1-1.8,1.2-3.3h1.9c-0.1,1.9-0.7,3.5-1.9,4.6
c-1.1,1.1-2.6,1.7-4.3,1.7c-2.3,0-4-0.7-5.1-2.2s-1.7-3.6-1.8-6.4v-2.3c0-2.9,0.6-5.1,1.7-6.6s2.9-2.2,5.1-2.2
c1.9,0,3.4,0.6,4.5,1.8s1.7,2.9,1.7,5H95c-0.1-1.6-0.5-2.8-1.2-3.6s-1.8-1.3-3.1-1.3c-1.7,0-2.9,0.6-3.7,1.7
c-0.8,1.1-1.2,2.9-1.2,5.2v2.2c0,2.4,0.4,4.2,1.2,5.3S89,122.4,90.6,122.4z"/>
<path class="st0" d="M100.5,113.4c0-2.7,0.6-4.9,1.9-6.5s2.9-2.4,5.1-2.4c2.2,0,3.9,0.8,5.1,2.4s1.9,3.7,1.9,6.5v2
c0,2.8-0.6,5-1.9,6.5s-2.9,2.3-5.1,2.3c-2.1,0-3.8-0.8-5-2.3s-1.9-3.6-1.9-6.3V113.4z M102.5,115.3c0,2.2,0.4,3.9,1.3,5.2
c0.9,1.3,2.1,1.9,3.7,1.9c1.6,0,2.8-0.6,3.7-1.8c0.8-1.2,1.3-2.9,1.3-5.2v-2c0-2.2-0.4-3.9-1.3-5.2c-0.9-1.3-2.1-1.9-3.7-1.9
c-1.5,0-2.7,0.6-3.6,1.8c-0.9,1.2-1.3,2.9-1.4,5.1V115.3z"/>
<path class="st0" d="M121.1,104.8l0.1,3c0.6-1,1.3-1.9,2.2-2.5s1.9-0.9,3-0.9c3.3,0,5,2.2,5.1,6.6v12.7h-1.9v-12.5
c0-1.7-0.3-3-0.9-3.8c-0.6-0.8-1.5-1.2-2.8-1.2c-1,0-1.9,0.4-2.8,1.2s-1.4,1.8-1.9,3.2v13.2h-2v-19H121.1z"/>
<path class="st0" d="M138.2,123.8v-17.3h-2.6v-1.8h2.6v-2.5c0-1.9,0.5-3.3,1.3-4.3s2-1.5,3.5-1.5c0.7,0,1.3,0.1,1.9,0.3
l-0.1,1.8c-0.5-0.1-1-0.2-1.6-0.2c-0.9,0-1.7,0.4-2.2,1.1s-0.8,1.7-0.8,3v2.3h3.7v1.8h-3.7v17.3H138.2z"/>
<path class="st0" d="M148,99.3c0-0.4,0.1-0.7,0.3-1s0.5-0.4,0.9-0.4s0.7,0.1,1,0.4s0.3,0.6,0.3,1s-0.1,0.7-0.3,1s-0.5,0.4-1,0.4
s-0.7-0.1-0.9-0.4S148,99.7,148,99.3z M150.2,123.8h-2v-19h2V123.8z"/>
<path class="st0" d="M155.3,113.5c0-3,0.5-5.2,1.5-6.7s2.6-2.2,4.6-2.2c2.2,0,3.8,1,4.9,3l0.1-2.7h1.8v19.6c0,2.3-0.6,4-1.6,5.2
c-1.1,1.2-2.7,1.8-4.7,1.8c-1,0-2.1-0.3-3.2-0.8s-1.9-1.1-2.4-1.9l0.9-1.4c1.3,1.5,2.8,2.2,4.5,2.2c1.6,0,2.7-0.4,3.5-1.3
c0.7-0.8,1.1-2.1,1.1-3.8v-3c-1.1,1.8-2.7,2.7-4.9,2.7c-2,0-3.5-0.7-4.5-2.2s-1.6-3.6-1.6-6.5V113.5z M157.2,115.2
c0,2.4,0.4,4.2,1.1,5.4s1.9,1.7,3.5,1.7c2.1,0,3.6-1,4.5-3v-9.7c-0.9-2.2-2.4-3.3-4.4-3.3c-1.6,0-2.8,0.6-3.5,1.7
s-1.1,2.9-1.1,5.3V115.2z"/>
</g>
</g>
<g>
<g>
<g>
<line class="st19" x1="58.9" y1="20.1" x2="77.9" y2="35.6"/>
<g>
<polygon class="st0" points="57.6,23.2 55.3,17.1 61.7,18.2 "/>
</g>
</g>
</g>
<circle class="st0" cx="48" cy="10.1" r="10.1"/>
<g>
<g>
<line class="st19" x1="77.9" y1="35.1" x2="96.2" y2="66.7"/>
<g>
<polygon class="st0" points="93,67.5 98.6,70.7 98.6,64.2 "/>
</g>
</g>
</g>
<g>
<g>
<line class="st19" x1="92.3" y1="38.9" x2="114.4" y2="43.6"/>
<g>
<polygon class="st0" points="92.6,42.3 87.8,38 93.9,36 "/>
</g>
</g>
</g>
<g>
<g>
<line class="st19" x1="77.9" y1="35.6" x2="94.5" y2="25.1"/>
<g>
<polygon class="st0" points="95.4,28.3 98.4,22.6 91.9,22.9 "/>
</g>
</g>
</g>
<g>
<g>
<line class="st19" x1="48.5" y1="41.3" x2="63.5" y2="38.7"/>
<g>
<polygon class="st0" points="63.1,42.1 68.1,38 62,35.7 "/>
</g>
</g>
</g>
<g>
<g>
<line class="st19" x1="111.4" y1="66.3" x2="124.4" y2="45.2"/>
<g>
<polygon class="st0" points="114.6,67.2 109,70.2 109.1,63.8 "/>
</g>
</g>
</g>
<g>
<g>
<line class="st19" x1="115.4" y1="30.4" x2="124.4" y2="45.2"/>
<g>
<polygon class="st0" points="113.2,32.8 113,26.4 118.7,29.5 "/>
</g>
</g>
</g>
<circle class="st0" cx="77.9" cy="35.6" r="10.1"/>
<circle class="st0" cx="44.6" cy="42.4" r="10.1"/>
<circle class="st0" cx="107.4" cy="18.1" r="10.1"/>
<circle class="st0" cx="104.1" cy="79.1" r="10.1"/>
<circle class="st0" cx="124.4" cy="45.2" r="10.1"/>
</g>
</g>
</g>
</svg>

After

Width:  |  Height:  |  Size: 6.2 KiB

View File

@@ -0,0 +1,132 @@
<?xml version="1.0" encoding="utf-8"?>
<svg version="1.1" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" x="0px" y="0px"
viewBox="0 0 260.4 71.4" style="enable-background:new 0 0 260.4 71.4;" xml:space="preserve">
<style type="text/css">
.st0{fill:#FFFFFF;}
.st1{fill:#E22434;}
.st2{display:none;}
.st3{fill:#1B3663;}
.st4{fill:#00B1D1;}
.st5{fill:#BFE6EF;}
.st6{fill:#69CBE0;}
.st7{fill:#0080BD;}
.st8{fill:#005DAB;}
.st9{fill:#183660;}
.st10{fill:none;stroke:#183660;stroke-width:1.9441;stroke-miterlimit:10;}
.st11{fill:none;stroke:#183660;stroke-width:1.6201;stroke-miterlimit:10;}
.st12{fill:none;stroke:#183660;stroke-width:1.2961;stroke-miterlimit:10;}
.st13{fill:none;stroke:#C0E6EF;stroke-width:1.9441;stroke-miterlimit:10;}
.st14{fill:#C0E6EF;}
.st15{fill:none;stroke:#C0E6EF;stroke-width:1.2961;stroke-miterlimit:10;}
.st16{fill:none;stroke:#C0E6EF;stroke-width:1.6201;stroke-miterlimit:10;}
.st17{fill:none;stroke:#FFFFFF;stroke-width:1.9441;stroke-miterlimit:10;}
.st18{fill:none;stroke:#FFFFFF;stroke-width:1.2961;stroke-miterlimit:10;}
.st19{fill:none;stroke:#FFFFFF;stroke-width:1.6201;stroke-miterlimit:10;}
</style>
<g id="Layer_2">
</g>
<g id="Layer_1">
<g>
<g>
<g>
<path class="st0" d="M96.7,26l0.1,1.8c1.1-1.4,2.6-2.1,4.4-2.1c1.9,0,3.2,0.9,4,2.6c1.1-1.7,2.6-2.6,4.7-2.6
c3.3,0,5,2.3,5.1,6.9V45h-5V32.9c0-1.1-0.2-1.9-0.5-2.4s-0.8-0.7-1.5-0.7c-0.9,0-1.6,0.6-2.1,1.7l0,0.6V45h-5V32.9
c0-1.1-0.1-1.9-0.4-2.4s-0.8-0.7-1.6-0.7c-0.9,0-1.5,0.5-2,1.4V45h-5V26H96.7z"/>
<path class="st0" d="M118.5,34.9c0-3.1,0.6-5.4,1.7-7s2.7-2.3,4.7-2.3c1.7,0,3.1,0.7,4,2l0.2-1.7h4.5v19c0,2.4-0.7,4.3-2,5.6
c-1.4,1.3-3.3,1.9-5.9,1.9c-1,0-2.1-0.2-3.3-0.6s-2-0.9-2.6-1.6l1.7-3.4c0.5,0.5,1.1,0.9,1.8,1.2c0.7,0.3,1.5,0.5,2.1,0.5
c1.1,0,1.9-0.3,2.4-0.8s0.7-1.4,0.7-2.6v-1.6c-0.9,1.2-2.2,1.9-3.7,1.9c-2,0-3.6-0.8-4.7-2.4s-1.7-3.8-1.7-6.7V34.9z
M123.5,36.2c0,1.8,0.2,3,0.7,3.8s1.2,1.2,2.2,1.2c1,0,1.8-0.4,2.3-1.1V31c-0.5-0.8-1.3-1.2-2.2-1.2c-1,0-1.7,0.4-2.2,1.2
s-0.7,2.1-0.7,3.9V36.2z"/>
<path class="st0" d="M142.2,26l0.1,1.8c1.1-1.4,2.6-2.1,4.4-2.1c1.9,0,3.2,0.9,4,2.6c1.1-1.7,2.6-2.6,4.7-2.6
c3.3,0,5,2.3,5.1,6.9V45h-5V32.9c0-1.1-0.2-1.9-0.5-2.4c-0.3-0.5-0.8-0.7-1.5-0.7c-0.9,0-1.6,0.6-2.1,1.7l0,0.6V45h-5V32.9
c0-1.1-0.1-1.9-0.4-2.4c-0.3-0.5-0.8-0.7-1.6-0.7c-0.9,0-1.5,0.5-2,1.4V45h-5V26H142.2z"/>
<path class="st0" d="M170.3,21.3V26h2.5v3.7h-2.5v9.5c0,0.8,0.1,1.3,0.3,1.5c0.2,0.3,0.6,0.4,1.2,0.4c0.5,0,0.9,0,1.2-0.1l0,3.9
c-0.8,0.3-1.8,0.5-2.7,0.5c-3.2,0-4.8-1.8-4.9-5.5V29.7h-2.2V26h2.2v-4.7H170.3z"/>
<path class="st0" d="M182.7,43.5c1.4,0,2.4-0.4,3.1-1.1c0.7-0.8,1.1-1.9,1.2-3.3h1.9c-0.1,1.9-0.7,3.5-1.9,4.6s-2.6,1.7-4.3,1.7
c-2.3,0-4-0.7-5.1-2.2s-1.7-3.6-1.8-6.4v-2.3c0-2.9,0.6-5.1,1.7-6.6s2.9-2.2,5.1-2.2c1.9,0,3.4,0.6,4.5,1.8s1.7,2.8,1.7,5H187
c-0.1-1.6-0.5-2.8-1.2-3.6s-1.8-1.3-3.1-1.3c-1.7,0-2.9,0.6-3.7,1.7c-0.8,1.1-1.2,2.9-1.2,5.2v2.2c0,2.4,0.4,4.2,1.2,5.3
C179.8,43,181,43.5,182.7,43.5z"/>
<path class="st0" d="M192.6,34.5c0-2.7,0.6-4.9,1.9-6.5s2.9-2.4,5.1-2.4c2.2,0,3.9,0.8,5.1,2.4c1.2,1.6,1.9,3.7,1.9,6.5v2
c0,2.8-0.6,5-1.9,6.5s-2.9,2.3-5.1,2.3s-3.8-0.8-5-2.3s-1.9-3.6-1.9-6.3V34.5z M194.6,36.5c0,2.2,0.4,3.9,1.3,5.2
c0.9,1.3,2.1,1.9,3.7,1.9c1.6,0,2.8-0.6,3.7-1.8c0.8-1.2,1.3-2.9,1.3-5.2v-2c0-2.2-0.4-3.9-1.3-5.2c-0.9-1.3-2.1-1.9-3.7-1.9
c-1.5,0-2.7,0.6-3.6,1.8c-0.9,1.2-1.3,2.9-1.4,5.1V36.5z"/>
<path class="st0" d="M213.2,26l0.1,3c0.6-1,1.3-1.9,2.2-2.5s1.9-0.9,3-0.9c3.3,0,5,2.2,5.1,6.6V45h-1.9V32.5
c0-1.7-0.3-3-0.9-3.8c-0.6-0.8-1.5-1.2-2.8-1.2c-1,0-1.9,0.4-2.8,1.2s-1.4,1.8-1.9,3.2V45h-2V26H213.2z"/>
<path class="st0" d="M230.3,45V27.7h-2.6V26h2.6v-2.5c0-1.9,0.5-3.3,1.3-4.3s2-1.5,3.5-1.5c0.7,0,1.3,0.1,1.9,0.3l-0.1,1.8
c-0.5-0.1-1-0.2-1.6-0.2c-0.9,0-1.7,0.4-2.2,1.1s-0.8,1.7-0.8,3V26h3.7v1.8h-3.7V45H230.3z"/>
<path class="st0" d="M240.1,20.5c0-0.4,0.1-0.7,0.3-1s0.5-0.4,0.9-0.4s0.7,0.1,1,0.4s0.3,0.6,0.3,1s-0.1,0.7-0.3,1
s-0.5,0.4-1,0.4s-0.7-0.1-0.9-0.4S240.1,20.9,240.1,20.5z M242.3,45h-2V26h2V45z"/>
<path class="st0" d="M247.4,34.6c0-3,0.5-5.2,1.5-6.7s2.6-2.2,4.6-2.2c2.2,0,3.8,1,4.9,3l0.1-2.7h1.8v19.6c0,2.3-0.6,4-1.6,5.2
s-2.7,1.8-4.7,1.8c-1,0-2.1-0.3-3.2-0.8s-1.9-1.1-2.4-1.9l0.9-1.4c1.3,1.5,2.8,2.2,4.5,2.2c1.6,0,2.7-0.4,3.5-1.3
c0.7-0.8,1.1-2.1,1.1-3.8v-3c-1.1,1.8-2.7,2.7-4.9,2.7c-2,0-3.5-0.7-4.5-2.2s-1.6-3.6-1.6-6.5V34.6z M249.3,36.4
c0,2.4,0.4,4.2,1.1,5.4s1.9,1.7,3.5,1.7c2.1,0,3.6-1,4.5-3v-9.7c-0.9-2.2-2.4-3.3-4.4-3.3c-1.6,0-2.8,0.6-3.5,1.7
s-1.1,2.9-1.1,5.3V36.4z"/>
</g>
</g>
<g>
<g>
<g>
<line class="st18" x1="19.5" y1="16" x2="34.7" y2="28.5"/>
<g>
<polygon class="st0" points="18.4,18.5 16.6,13.7 21.7,14.5 "/>
</g>
</g>
</g>
<circle class="st0" cx="10.8" cy="8.1" r="8.1"/>
<g>
<g>
<line class="st18" x1="34.7" y1="28" x2="49.4" y2="53.3"/>
<g>
<polygon class="st0" points="46.8,54 51.2,56.5 51.2,51.4 "/>
</g>
</g>
</g>
<g>
<g>
<line class="st18" x1="46.2" y1="31.1" x2="63.9" y2="34.9"/>
<g>
<polygon class="st0" points="46.4,33.8 42.6,30.4 47.5,28.8 "/>
</g>
</g>
</g>
<g>
<g>
<line class="st18" x1="34.7" y1="28.5" x2="47.9" y2="20.1"/>
<g>
<polygon class="st0" points="48.7,22.7 51.1,18.1 45.9,18.3 "/>
</g>
</g>
</g>
<g>
<g>
<line class="st18" x1="11.2" y1="33.1" x2="23.2" y2="31"/>
<g>
<polygon class="st0" points="22.9,33.7 26.8,30.4 22,28.6 "/>
</g>
</g>
</g>
<g>
<g>
<line class="st18" x1="61.5" y1="53" x2="71.9" y2="36.1"/>
<g>
<polygon class="st0" points="64.1,53.7 59.5,56.2 59.7,51 "/>
</g>
</g>
</g>
<g>
<g>
<line class="st18" x1="64.7" y1="24.4" x2="71.9" y2="36.6"/>
<g>
<polygon class="st0" points="62.9,26.4 62.9,21.2 67.3,23.7 "/>
</g>
</g>
</g>
<circle class="st0" cx="34.7" cy="28.5" r="8.1"/>
<circle class="st0" cx="8.1" cy="33.9" r="8.1"/>
<circle class="st0" cx="58.3" cy="14.5" r="8.1"/>
<circle class="st0" cx="55.7" cy="63.3" r="8.1"/>
<circle class="st0" cx="71.9" cy="36.1" r="8.1"/>
</g>
</g>
</g>
</svg>

After

Width:  |  Height:  |  Size: 6.2 KiB

BIN
art/mgmt_poohbear_meme.jpg Normal file

Binary file not shown.

After

Width:  |  Height:  |  Size: 102 KiB

189
cli/cli.go Normal file
View File

@@ -0,0 +1,189 @@
// Mgmt
// Copyright (C) James Shubin and the project contributors
// Written by James Shubin <james@shubin.ca> and the project contributors
//
// This program is free software: you can redistribute it and/or modify
// it under the terms of the GNU General Public License as published by
// the Free Software Foundation, either version 3 of the License, or
// (at your option) any later version.
//
// This program is distributed in the hope that it will be useful,
// but WITHOUT ANY WARRANTY; without even the implied warranty of
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
// GNU General Public License for more details.
//
// You should have received a copy of the GNU General Public License
// along with this program. If not, see <https://www.gnu.org/licenses/>.
//
// Additional permission under GNU GPL version 3 section 7
//
// If you modify this program, or any covered work, by linking or combining it
// with embedded mcl code and modules (and that the embedded mcl code and
// modules which link with this program, contain a copy of their source code in
// the authoritative form) containing parts covered by the terms of any other
// license, the licensors of this program grant you additional permission to
// convey the resulting work. Furthermore, the licensors of this program grant
// the original author, James Shubin, additional permission to update this
// additional permission if he deems it necessary to achieve the goals of this
// additional permission.
// Package cli handles all of the core command line parsing. It's the first
// entry point after the real main function, and it imports and runs our core
// "lib".
package cli
import (
"context"
"fmt"
"os"
cliUtil "github.com/purpleidea/mgmt/cli/util"
"github.com/purpleidea/mgmt/util/errwrap"
"github.com/alexflint/go-arg"
)
func init() {
if _, err := arg.NewParser(arg.Config{}, &Args{}); err != nil { // sanity check
panic(errwrap.Wrapf(err, "invalid args cli struct"))
}
}
// CLI is the entry point for using mgmt normally from the CLI.
func CLI(ctx context.Context, data *cliUtil.Data) error {
// test for sanity
if data == nil {
return fmt.Errorf("this CLI was not run correctly")
}
if data.Program == "" || data.Version == "" {
return fmt.Errorf("program was not compiled correctly")
}
if data.Copying == "" {
return fmt.Errorf("program copyrights were removed, can't run")
}
args := Args{}
args.version = data.Version // copy this in
args.description = data.Tagline
config := arg.Config{
Program: data.Program,
}
parser, err := arg.NewParser(config, &args)
if err != nil {
// programming error
return errwrap.Wrapf(err, "cli config error")
}
err = parser.Parse(data.Args[1:]) // XXX: args[0] needs to be dropped
if err == arg.ErrHelp {
parser.WriteHelp(os.Stdout)
return nil
}
if err == arg.ErrVersion {
fmt.Printf("%s\n", data.Version) // byon: bring your own newline
return nil
}
if err != nil {
//parser.WriteHelp(os.Stdout) // TODO: is doing this helpful?
return cliUtil.CliParseError(err) // consistent errors
}
// display the license
if args.License {
fmt.Printf("%s", data.Copying) // file comes with a trailing nl
return nil
}
if ok, err := args.Run(ctx, data); err != nil {
return err
} else if ok { // did we activate one of the commands?
return nil
}
// print help if no subcommands are set
parser.WriteHelp(os.Stdout)
return nil
}
// Args is the CLI parsing structure and type of the parsed result. This
// particular struct is the top-most one.
type Args struct {
// XXX: We cannot have both subcommands and a positional argument.
// XXX: I think it's a bug of this library that it can't handle argv[0].
//Argv0 string `arg:"positional"`
License bool `arg:"--license" help:"display the license and exit"`
RunCmd *RunArgs `arg:"subcommand:run" help:"run code on this machine"`
DeployCmd *DeployArgs `arg:"subcommand:deploy" help:"deploy code into a cluster"`
SetupCmd *SetupArgs `arg:"subcommand:setup" help:"setup some bootstrapping tasks"`
FirstbootCmd *FirstbootArgs `arg:"subcommand:firstboot" help:"run some tasks on first boot"`
DocsCmd *DocsGenerateArgs `arg:"subcommand:docs" help:"generate documentation"`
ToolsCmd *ToolsArgs `arg:"subcommand:tools" help:"collection of useful tools"`
// This never runs, it gets preempted in the real main() function.
// XXX: Can we do it nicely with the new arg parser? can it ignore all args?
EtcdCmd *EtcdArgs `arg:"subcommand:etcd" help:"run standalone etcd"`
// version is a private handle for our version string.
version string `arg:"-"` // ignored from parsing
// description is a private handle for our description string.
description string `arg:"-"` // ignored from parsing
}
// Version returns the version string. Implementing this signature is part of
// the API for the cli library.
func (obj *Args) Version() string {
return obj.version
}
// Description returns a description string. Implementing this signature is part
// of the API for the cli library.
func (obj *Args) Description() string {
return obj.description
}
// Run executes the correct subcommand. It errors if there's ever an error. It
// returns true if we did activate one of the subcommands. It returns false if
// we did not. This information is used so that the top-level parser can return
// usage or help information if no subcommand activates.
func (obj *Args) Run(ctx context.Context, data *cliUtil.Data) (bool, error) {
if cmd := obj.RunCmd; cmd != nil {
return cmd.Run(ctx, data)
}
if cmd := obj.DeployCmd; cmd != nil {
return cmd.Run(ctx, data)
}
if cmd := obj.SetupCmd; cmd != nil {
return cmd.Run(ctx, data)
}
if cmd := obj.FirstbootCmd; cmd != nil {
return cmd.Run(ctx, data)
}
if cmd := obj.DocsCmd; cmd != nil {
return cmd.Run(ctx, data)
}
if cmd := obj.ToolsCmd; cmd != nil {
return cmd.Run(ctx, data)
}
// NOTE: we could return true, fmt.Errorf("...") if more than one did
return false, nil // nobody activated
}
// EtcdArgs is the CLI parsing structure and type of the parsed result. This
// particular one is empty because the `etcd` subcommand is preempted in the
// real main() function.
type EtcdArgs struct{}

314
cli/deploy.go Normal file
View File

@@ -0,0 +1,314 @@
// Mgmt
// Copyright (C) James Shubin and the project contributors
// Written by James Shubin <james@shubin.ca> and the project contributors
//
// This program is free software: you can redistribute it and/or modify
// it under the terms of the GNU General Public License as published by
// the Free Software Foundation, either version 3 of the License, or
// (at your option) any later version.
//
// This program is distributed in the hope that it will be useful,
// but WITHOUT ANY WARRANTY; without even the implied warranty of
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
// GNU General Public License for more details.
//
// You should have received a copy of the GNU General Public License
// along with this program. If not, see <https://www.gnu.org/licenses/>.
//
// Additional permission under GNU GPL version 3 section 7
//
// If you modify this program, or any covered work, by linking or combining it
// with embedded mcl code and modules (and that the embedded mcl code and
// modules which link with this program, contain a copy of their source code in
// the authoritative form) containing parts covered by the terms of any other
// license, the licensors of this program grant you additional permission to
// convey the resulting work. Furthermore, the licensors of this program grant
// the original author, James Shubin, additional permission to update this
// additional permission if he deems it necessary to achieve the goals of this
// additional permission.
package cli
import (
"context"
"fmt"
"os"
"os/signal"
cliUtil "github.com/purpleidea/mgmt/cli/util"
"github.com/purpleidea/mgmt/engine"
"github.com/purpleidea/mgmt/etcd"
"github.com/purpleidea/mgmt/etcd/client"
etcdfs "github.com/purpleidea/mgmt/etcd/fs"
etcdSSH "github.com/purpleidea/mgmt/etcd/ssh"
"github.com/purpleidea/mgmt/gapi"
"github.com/purpleidea/mgmt/lib"
"github.com/purpleidea/mgmt/util"
"github.com/purpleidea/mgmt/util/errwrap"
git "github.com/go-git/go-git/v5"
"github.com/google/uuid"
)
// DeployArgs is the CLI parsing structure and type of the parsed result. This
// particular one contains all the common flags for the `deploy` subcommand
// which all frontends can use.
type DeployArgs struct {
// SSHURL can be specified if we want to transport the SSH client
// connection over SSH. If this is specified, the second hop is made
// with the Seeds values, but they connect from this destination. You
// can specify this in the standard james@server:22 format. This will
// use your ~/.ssh/ directory for public key authentication and
// verifying the host key in the known_hosts file. This must already be
// setup for things to work.
SSHURL string `arg:"--ssh-url" help:"transport the etcd client connection over SSH to this server"`
// SSHHostKey is the key part (which is already base64 encoded) from a
// known_hosts file, representing the host we're connecting to. If this
// is specified, then it overrides looking for it in the URL.
SSHHostKey string `arg:"--ssh-hostkey" help:"use this ssh known hosts key when connecting over SSH"`
Seeds []string `arg:"--seeds,separate,env:MGMT_SEEDS" help:"default etcd client endpoints"`
Noop bool `arg:"--noop" help:"globally force all resources into no-op mode"`
Sema int `arg:"--sema" default:"-1" help:"globally add a semaphore to all resources with this lock count"`
NoGit bool `arg:"--no-git" help:"don't look at git commit id for safe deploys"`
Force bool `arg:"--force" help:"force a new deploy, even if the safety chain would break"`
NoAutoEdges bool `arg:"--no-autoedges" help:"skip the autoedges stage"`
DeployEmpty *cliUtil.EmptyArgs `arg:"subcommand:empty" help:"deploy empty payload"`
DeployLang *cliUtil.LangArgs `arg:"subcommand:lang" help:"deploy lang (mcl) payload"`
DeployYaml *cliUtil.YamlArgs `arg:"subcommand:yaml" help:"deploy yaml graph payload"`
DeployPuppet *cliUtil.PuppetArgs `arg:"subcommand:puppet" help:"deploy puppet graph payload"`
DeployLangPuppet *cliUtil.LangPuppetArgs `arg:"subcommand:langpuppet" help:"deploy langpuppet graph payload"`
}
// Run executes the correct subcommand. It errors if there's ever an error. It
// returns true if we did activate one of the subcommands. It returns false if
// we did not. This information is used so that the top-level parser can return
// usage or help information if no subcommand activates. This particular Run is
// the run for the main `deploy` subcommand. This always requires a frontend to
// deploy to the cluster, but if you don't want a graph, you can use the `empty`
// frontend. The engine backend is agnostic to which frontend is deployed, in
// fact, you can deploy with multiple different frontends, one after another, on
// the same engine.
func (obj *DeployArgs) Run(ctx context.Context, data *cliUtil.Data) (bool, error) {
var name string
var args interface{}
if cmd := obj.DeployEmpty; cmd != nil {
name = cliUtil.LookupSubcommand(obj, cmd) // "empty"
args = cmd
}
if cmd := obj.DeployLang; cmd != nil {
name = cliUtil.LookupSubcommand(obj, cmd) // "lang"
args = cmd
}
if cmd := obj.DeployYaml; cmd != nil {
name = cliUtil.LookupSubcommand(obj, cmd) // "yaml"
args = cmd
}
if cmd := obj.DeployPuppet; cmd != nil {
name = cliUtil.LookupSubcommand(obj, cmd) // "puppet"
args = cmd
}
if cmd := obj.DeployLangPuppet; cmd != nil {
name = cliUtil.LookupSubcommand(obj, cmd) // "langpuppet"
args = cmd
}
// XXX: workaround https://github.com/alexflint/go-arg/issues/239
gapiNames := gapi.Names() // list of registered names
if l := len(obj.Seeds); name == "" && l > 1 {
elem := obj.Seeds[l-2] // second to last element
if util.StrInList(elem, gapiNames) {
return false, cliUtil.CliParseError(cliUtil.MissingEquals) // consistent errors
}
}
fn, exists := gapi.RegisteredGAPIs[name]
if !exists {
return false, nil // did not activate
}
gapiObj := fn()
program, version := data.Program, data.Version
Logf := func(format string, v ...interface{}) {
data.Flags.Logf("deploy: "+format, v...)
}
// TODO: consider adding a timeout based on an args.Timeout flag ?
ctx, cancel := signal.NotifyContext(ctx, os.Interrupt)
defer cancel()
cliUtil.Hello(program, version, data.Flags) // say hello!
defer Logf("goodbye!")
var hash, pHash string
if !obj.NoGit {
wd, err := os.Getwd()
if err != nil {
return false, errwrap.Wrapf(err, "could not get current working directory")
}
repo, err := git.PlainOpen(wd)
if err != nil {
return false, errwrap.Wrapf(err, "could not open git repo")
}
head, err := repo.Head()
if err != nil {
return false, errwrap.Wrapf(err, "could not read git HEAD")
}
hash = head.Hash().String() // current commit id
Logf("hash: %s", hash)
lo := &git.LogOptions{
From: head.Hash(),
}
commits, err := repo.Log(lo)
if err != nil {
return false, errwrap.Wrapf(err, "could not read git log")
}
if _, err := commits.Next(); err != nil { // skip over HEAD
return false, errwrap.Wrapf(err, "could not read HEAD in git log") // weird!
}
commit, err := commits.Next()
if err == nil { // errors are okay, we might be empty
pHash = commit.Hash.String() // previous commit id
}
Logf("previous deploy hash: %s", pHash)
if obj.Force {
pHash = "" // don't check this :(
}
if hash == "" {
return false, errwrap.Wrapf(err, "could not get git deploy hash")
}
}
uniqueid := uuid.New() // panic's if it can't generate one :P
etcdClient := client.NewClientFromSeedsNamespace(
obj.Seeds, // endpoints
lib.NS,
)
if err := etcdClient.Init(); err != nil {
return false, errwrap.Wrapf(err, "client Init failed")
}
defer func() {
err := errwrap.Wrapf(etcdClient.Close(), "client Close failed")
if err != nil {
// TODO: cause the final exit code to be non-zero
Logf("client cleanup error: %+v", err)
}
}()
var world engine.World
world = &etcd.World{ // XXX: What should some of these fields be?
Client: etcdClient, // XXX: remove me when etcdfs below is done
Seeds: obj.Seeds,
NS: lib.NS,
//MetadataPrefix: lib.MetadataPrefix,
//StoragePrefix: lib.StoragePrefix,
//StandaloneFs: ???.DeployFs, // used for static deploys
//GetURI: func() string {
//},
}
if obj.SSHURL != "" { // alternate world implementation over SSH
world = &etcdSSH.World{
URL: obj.SSHURL,
HostKey: obj.SSHHostKey,
Seeds: obj.Seeds,
NS: lib.NS,
//MetadataPrefix: lib.MetadataPrefix,
//StoragePrefix: lib.StoragePrefix,
//StandaloneFs: ???.DeployFs, // used for static deploys
//GetURI: func() string {
//},
}
// XXX: We need to first get rid of the standalone etcd client,
// and then pull the etcdfs stuff in so it uses that client.
return false, fmt.Errorf("--ssh-url is not implemented yet")
}
worldInit := &engine.WorldInit{
Hostname: "", // XXX: Should we set this?
Debug: data.Flags.Debug,
Logf: func(format string, v ...interface{}) {
Logf("world: etcd: "+format, v...)
},
}
if err := world.Connect(ctx, worldInit); err != nil {
return false, errwrap.Wrapf(err, "world Connect failed")
}
defer func() {
err := errwrap.Wrapf(world.Cleanup(), "world Cleanup failed")
if err != nil {
// TODO: cause the final exit code to be non-zero?
Logf("close error: %+v", err)
}
}()
// get max id (from all the previous deploys)
max, err := world.GetMaxDeployID(ctx)
if err != nil {
return false, errwrap.Wrapf(err, "error getting max deploy id")
}
// find the latest id
var id = max + 1 // next id
Logf("previous max deploy id: %d", max)
// XXX: Get this from the World API? (Which might need improving!)
etcdFs := &etcdfs.Fs{
Client: etcdClient,
// TODO: using a uuid is meant as a temporary measure, i hate them
Metadata: lib.MetadataPrefix + fmt.Sprintf("/deploy/%d-%s", id, uniqueid),
DataPrefix: lib.StoragePrefix,
Debug: data.Flags.Debug,
Logf: func(format string, v ...interface{}) {
Logf("fs: "+format, v...)
},
}
info := &gapi.Info{
Args: args,
Flags: &gapi.Flags{
Noop: obj.Noop,
Sema: obj.Sema,
//Update: obj.Update,
},
Fs: etcdFs,
Debug: data.Flags.Debug,
Logf: func(format string, v ...interface{}) {
// TODO: is this a sane prefix to use here?
data.Flags.Logf("cli: "+format, v...)
},
}
deploy, err := gapiObj.Cli(info)
if err != nil {
return false, cliUtil.CliParseError(err) // consistent errors
}
if deploy == nil { // not used
return false, fmt.Errorf("not enough information specified")
}
// redundant
deploy.Noop = obj.Noop
deploy.Sema = obj.Sema
deploy.NoAutoEdges = obj.NoAutoEdges
str, err := deploy.ToB64()
if err != nil {
return false, errwrap.Wrapf(err, "encoding error")
}
Logf("pushing...")
// this nominally checks the previous git hash matches our expectation
if err := world.AddDeploy(ctx, id, hash, pHash, &str); err != nil {
return false, errwrap.Wrapf(err, "could not create deploy id `%d`", id)
}
Logf("success, id: %d", id)
return true, nil
}

150
cli/docs.go Normal file
View File

@@ -0,0 +1,150 @@
// Mgmt
// Copyright (C) James Shubin and the project contributors
// Written by James Shubin <james@shubin.ca> and the project contributors
//
// This program is free software: you can redistribute it and/or modify
// it under the terms of the GNU General Public License as published by
// the Free Software Foundation, either version 3 of the License, or
// (at your option) any later version.
//
// This program is distributed in the hope that it will be useful,
// but WITHOUT ANY WARRANTY; without even the implied warranty of
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
// GNU General Public License for more details.
//
// You should have received a copy of the GNU General Public License
// along with this program. If not, see <https://www.gnu.org/licenses/>.
//
// Additional permission under GNU GPL version 3 section 7
//
// If you modify this program, or any covered work, by linking or combining it
// with embedded mcl code and modules (and that the embedded mcl code and
// modules which link with this program, contain a copy of their source code in
// the authoritative form) containing parts covered by the terms of any other
// license, the licensors of this program grant you additional permission to
// convey the resulting work. Furthermore, the licensors of this program grant
// the original author, James Shubin, additional permission to update this
// additional permission if he deems it necessary to achieve the goals of this
// additional permission.
package cli
import (
"context"
"os"
"os/signal"
"sync"
"syscall"
cliUtil "github.com/purpleidea/mgmt/cli/util"
"github.com/purpleidea/mgmt/docs"
)
// DocsGenerateArgs is the CLI parsing structure and type of the parsed result.
// This particular one contains all the common flags for the `docs generate`
// subcommand.
type DocsGenerateArgs struct {
docs.Config // embedded config (can't be a pointer) https://github.com/alexflint/go-arg/issues/240
DocsGenerate *cliUtil.DocsGenerateArgs `arg:"subcommand:generate" help:"generate documentation"`
}
// Run executes the correct subcommand. It errors if there's ever an error. It
// returns true if we did activate one of the subcommands. It returns false if
// we did not. This information is used so that the top-level parser can return
// usage or help information if no subcommand activates. This particular Run is
// the run for the main `docs` subcommand.
func (obj *DocsGenerateArgs) Run(ctx context.Context, data *cliUtil.Data) (bool, error) {
ctx, cancel := context.WithCancel(ctx)
defer cancel()
var name string
var args interface{}
if cmd := obj.DocsGenerate; cmd != nil {
name = cliUtil.LookupSubcommand(obj, cmd) // "generate"
args = cmd
}
_ = name
Logf := func(format string, v ...interface{}) {
// Don't block this globally...
//if !data.Flags.Debug {
// return
//}
data.Flags.Logf("main: "+format, v...)
}
var api docs.API
if cmd := obj.DocsGenerate; cmd != nil {
api = &docs.Generate{
DocsGenerateArgs: args.(*cliUtil.DocsGenerateArgs),
Config: obj.Config,
Program: data.Program,
Version: data.Version,
Debug: data.Flags.Debug,
Logf: Logf,
}
}
if api == nil {
return false, nil // nothing found (display help!)
}
// We don't use these for the setup command in normal operation.
if data.Flags.Debug {
cliUtil.Hello(data.Program, data.Version, data.Flags) // say hello!
defer Logf("goodbye!")
}
// install the exit signal handler
wg := &sync.WaitGroup{}
defer wg.Wait()
exit := make(chan struct{})
defer close(exit)
wg.Add(1)
go func() {
defer cancel()
defer wg.Done()
// must have buffer for max number of signals
signals := make(chan os.Signal, 3+1) // 3 * ^C + 1 * SIGTERM
signal.Notify(signals, os.Interrupt) // catch ^C
//signal.Notify(signals, os.Kill) // catch signals
signal.Notify(signals, syscall.SIGTERM)
var count uint8
for {
select {
case sig := <-signals: // any signal will do
if sig != os.Interrupt {
data.Flags.Logf("interrupted by signal")
return
}
switch count {
case 0:
data.Flags.Logf("interrupted by ^C")
cancel()
case 1:
data.Flags.Logf("interrupted by ^C (fast pause)")
cancel()
case 2:
data.Flags.Logf("interrupted by ^C (hard interrupt)")
cancel()
}
count++
case <-exit:
return
}
}
}()
if err := api.Main(ctx); err != nil {
if data.Flags.Debug {
data.Flags.Logf("main: %+v", err)
}
return false, err
}
return true, nil
}

151
cli/firstboot.go Normal file
View File

@@ -0,0 +1,151 @@
// Mgmt
// Copyright (C) James Shubin and the project contributors
// Written by James Shubin <james@shubin.ca> and the project contributors
//
// This program is free software: you can redistribute it and/or modify
// it under the terms of the GNU General Public License as published by
// the Free Software Foundation, either version 3 of the License, or
// (at your option) any later version.
//
// This program is distributed in the hope that it will be useful,
// but WITHOUT ANY WARRANTY; without even the implied warranty of
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
// GNU General Public License for more details.
//
// You should have received a copy of the GNU General Public License
// along with this program. If not, see <https://www.gnu.org/licenses/>.
//
// Additional permission under GNU GPL version 3 section 7
//
// If you modify this program, or any covered work, by linking or combining it
// with embedded mcl code and modules (and that the embedded mcl code and
// modules which link with this program, contain a copy of their source code in
// the authoritative form) containing parts covered by the terms of any other
// license, the licensors of this program grant you additional permission to
// convey the resulting work. Furthermore, the licensors of this program grant
// the original author, James Shubin, additional permission to update this
// additional permission if he deems it necessary to achieve the goals of this
// additional permission.
package cli
import (
"context"
"os"
"os/signal"
"sync"
"syscall"
cliUtil "github.com/purpleidea/mgmt/cli/util"
"github.com/purpleidea/mgmt/firstboot"
)
// FirstbootArgs is the CLI parsing structure and type of the parsed result.
// This particular one contains all the common flags for the `firstboot`
// subcommand.
type FirstbootArgs struct {
firstboot.Config // embedded config (can't be a pointer) https://github.com/alexflint/go-arg/issues/240
FirstbootStart *cliUtil.FirstbootStartArgs `arg:"subcommand:start" help:"start firstboot service"`
}
// Run executes the correct subcommand. It errors if there's ever an error. It
// returns true if we did activate one of the subcommands. It returns false if
// we did not. This information is used so that the top-level parser can return
// usage or help information if no subcommand activates. This particular Run is
// the run for the main `firstboot` subcommand. The firstboot command as a
// service that lets you run commands once on the first boot of a system.
func (obj *FirstbootArgs) Run(ctx context.Context, data *cliUtil.Data) (bool, error) {
ctx, cancel := context.WithCancel(ctx)
defer cancel()
var name string
var args interface{}
if cmd := obj.FirstbootStart; cmd != nil {
name = cliUtil.LookupSubcommand(obj, cmd) // "pkg"
args = cmd
}
_ = name
Logf := func(format string, v ...interface{}) {
// Don't block this globally...
//if !data.Flags.Debug {
// return
//}
data.Flags.Logf("main: "+format, v...)
}
var api firstboot.API
if cmd := obj.FirstbootStart; cmd != nil {
api = &firstboot.Start{
FirstbootStartArgs: args.(*cliUtil.FirstbootStartArgs),
Config: obj.Config,
Program: data.Program,
Version: data.Version,
Debug: data.Flags.Debug,
Logf: Logf,
}
}
if api == nil {
return false, nil // nothing found (display help!)
}
// We don't use these for the setup command in normal operation.
if data.Flags.Debug {
cliUtil.Hello(data.Program, data.Version, data.Flags) // say hello!
defer Logf("goodbye!")
}
// install the exit signal handler
wg := &sync.WaitGroup{}
defer wg.Wait()
exit := make(chan struct{})
defer close(exit)
wg.Add(1)
go func() {
defer cancel()
defer wg.Done()
// must have buffer for max number of signals
signals := make(chan os.Signal, 3+1) // 3 * ^C + 1 * SIGTERM
signal.Notify(signals, os.Interrupt) // catch ^C
//signal.Notify(signals, os.Kill) // catch signals
signal.Notify(signals, syscall.SIGTERM)
var count uint8
for {
select {
case sig := <-signals: // any signal will do
if sig != os.Interrupt {
data.Flags.Logf("interrupted by signal")
return
}
switch count {
case 0:
data.Flags.Logf("interrupted by ^C")
cancel()
case 1:
data.Flags.Logf("interrupted by ^C (fast pause)")
cancel()
case 2:
data.Flags.Logf("interrupted by ^C (hard interrupt)")
cancel()
}
count++
case <-exit:
return
}
}
}()
if err := api.Main(ctx); err != nil {
if data.Flags.Debug {
data.Flags.Logf("main: %+v", err)
}
return false, err
}
return true, nil
}

245
cli/run.go Normal file
View File

@@ -0,0 +1,245 @@
// Mgmt
// Copyright (C) James Shubin and the project contributors
// Written by James Shubin <james@shubin.ca> and the project contributors
//
// This program is free software: you can redistribute it and/or modify
// it under the terms of the GNU General Public License as published by
// the Free Software Foundation, either version 3 of the License, or
// (at your option) any later version.
//
// This program is distributed in the hope that it will be useful,
// but WITHOUT ANY WARRANTY; without even the implied warranty of
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
// GNU General Public License for more details.
//
// You should have received a copy of the GNU General Public License
// along with this program. If not, see <https://www.gnu.org/licenses/>.
//
// Additional permission under GNU GPL version 3 section 7
//
// If you modify this program, or any covered work, by linking or combining it
// with embedded mcl code and modules (and that the embedded mcl code and
// modules which link with this program, contain a copy of their source code in
// the authoritative form) containing parts covered by the terms of any other
// license, the licensors of this program grant you additional permission to
// convey the resulting work. Furthermore, the licensors of this program grant
// the original author, James Shubin, additional permission to update this
// additional permission if he deems it necessary to achieve the goals of this
// additional permission.
package cli
import (
"context"
"fmt"
"os"
"os/signal"
"sync"
"syscall"
cliUtil "github.com/purpleidea/mgmt/cli/util"
"github.com/purpleidea/mgmt/gapi"
"github.com/purpleidea/mgmt/lib"
"github.com/purpleidea/mgmt/util"
"github.com/purpleidea/mgmt/util/errwrap"
"github.com/spf13/afero"
)
// RunArgs is the CLI parsing structure and type of the parsed result. This
// particular one contains all the common flags for the `run` subcommand which
// all frontends can use.
type RunArgs struct {
lib.Config // embedded config (can't be a pointer) https://github.com/alexflint/go-arg/issues/240
RunEmpty *cliUtil.EmptyArgs `arg:"subcommand:empty" help:"run empty payload"`
RunLang *cliUtil.LangArgs `arg:"subcommand:lang" help:"run lang (mcl) payload"`
RunYaml *cliUtil.YamlArgs `arg:"subcommand:yaml" help:"run yaml graph payload"`
RunPuppet *cliUtil.PuppetArgs `arg:"subcommand:puppet" help:"run puppet graph payload"`
RunLangPuppet *cliUtil.LangPuppetArgs `arg:"subcommand:langpuppet" help:"run a combined lang/puppet graph payload"`
}
// Run executes the correct subcommand. It errors if there's ever an error. It
// returns true if we did activate one of the subcommands. It returns false if
// we did not. This information is used so that the top-level parser can return
// usage or help information if no subcommand activates. This particular Run is
// the run for the main `run` subcommand. This always requires a frontend to
// start the engine, but if you don't want a graph, you can use the `empty`
// frontend. The engine backend is agnostic to which frontend is running, in
// fact, you can deploy with multiple different frontends, one after another, on
// the same engine.
func (obj *RunArgs) Run(ctx context.Context, data *cliUtil.Data) (bool, error) {
var name string
var args interface{}
if cmd := obj.RunEmpty; cmd != nil {
name = cliUtil.LookupSubcommand(obj, cmd) // "empty"
args = cmd
}
if cmd := obj.RunLang; cmd != nil {
name = cliUtil.LookupSubcommand(obj, cmd) // "lang"
args = cmd
}
if cmd := obj.RunYaml; cmd != nil {
name = cliUtil.LookupSubcommand(obj, cmd) // "yaml"
args = cmd
}
if cmd := obj.RunPuppet; cmd != nil {
name = cliUtil.LookupSubcommand(obj, cmd) // "puppet"
args = cmd
}
if cmd := obj.RunLangPuppet; cmd != nil {
name = cliUtil.LookupSubcommand(obj, cmd) // "langpuppet"
args = cmd
}
// XXX: workaround https://github.com/alexflint/go-arg/issues/239
lists := [][]string{
obj.Seeds,
obj.ClientURLs,
obj.ServerURLs,
obj.AdvertiseClientURLs,
obj.AdvertiseServerURLs,
}
gapiNames := gapi.Names() // list of registered names
for _, list := range lists {
if l := len(list); name == "" && l > 1 {
elem := list[l-2] // second to last element
if util.StrInList(elem, gapiNames) {
return false, cliUtil.CliParseError(cliUtil.MissingEquals) // consistent errors
}
}
}
fn, exists := gapi.RegisteredGAPIs[name]
if !exists {
return false, nil // did not activate
}
gapiObj := fn()
main := &lib.Main{}
main.Config = &obj.Config // pass in all the parsed data
main.Program, main.Version = data.Program, data.Version
main.Debug, main.Logf = data.Flags.Debug, data.Flags.Logf // no prefix
Logf := func(format string, v ...interface{}) {
data.Flags.Logf("main: "+format, v...)
}
cliUtil.Hello(main.Program, main.Version, data.Flags) // say hello!
defer Logf("goodbye!")
// create a memory backed temporary filesystem for storing runtime data
mmFs := afero.NewMemMapFs()
afs := &afero.Afero{Fs: mmFs} // wrap so that we're implementing ioutil
standaloneFs := &util.AferoFs{Afero: afs}
main.DeployFs = standaloneFs
info := &gapi.Info{
Args: args,
Flags: &gapi.Flags{
Hostname: obj.Hostname,
Noop: obj.Noop,
Sema: obj.Sema,
//Update: obj.Update,
NoAutoEdges: obj.NoAutoEdges,
},
Fs: standaloneFs,
Debug: data.Flags.Debug,
Logf: func(format string, v ...interface{}) {
data.Flags.Logf("cli: "+format, v...)
},
}
deploy, err := gapiObj.Cli(info)
if err != nil {
return false, cliUtil.CliParseError(err) // consistent errors
}
if cmd := obj.RunLang; cmd != nil && cmd.OnlyUnify && deploy == nil {
return true, nil // we end early
}
if cmd := obj.RunLang; cmd != nil && cmd.OnlyDownload && deploy == nil {
return true, nil // we end early
}
main.Deploy = deploy
if main.Deploy == nil {
// nobody activated, but we'll still watch the etcd deploy chan,
// and if there is deployed code that's ready to run, we'll run!
data.Flags.Logf("main: no frontend selected (no GAPI activated)")
}
if err := main.Validate(); err != nil {
return false, err
}
if err := main.Init(); err != nil {
return false, err
}
// install the exit signal handler
wg := &sync.WaitGroup{}
defer wg.Wait()
exit := make(chan struct{})
defer close(exit)
wg.Add(1)
go func() {
defer wg.Done()
// must have buffer for max number of signals
signals := make(chan os.Signal, 3+1) // 3 * ^C + 1 * SIGTERM
signal.Notify(signals, os.Interrupt) // catch ^C
//signal.Notify(signals, os.Kill) // catch signals
signal.Notify(signals, syscall.SIGTERM)
var count uint8
for {
select {
case sig := <-signals: // any signal will do
if sig != os.Interrupt {
data.Flags.Logf("interrupted by signal")
main.Interrupt(fmt.Errorf("killed by %v", sig))
return
}
switch count {
case 0:
data.Flags.Logf("interrupted by ^C")
main.Exit(nil)
case 1:
data.Flags.Logf("interrupted by ^C (fast pause)")
main.FastExit(nil)
case 2:
data.Flags.Logf("interrupted by ^C (hard interrupt)")
main.Interrupt(nil)
}
count++
case <-exit:
return
}
}
}()
reterr := main.Run()
if reterr != nil {
// log the error message returned
if data.Flags.Debug {
data.Flags.Logf("main: %+v", reterr)
}
}
if err := main.Close(); err != nil {
if data.Flags.Debug {
data.Flags.Logf("main: Close: %+v", err)
}
if reterr == nil {
return false, err
}
reterr = errwrap.Append(reterr, err)
}
if reterr != nil {
return false, reterr
}
return true, nil
}

180
cli/setup.go Normal file
View File

@@ -0,0 +1,180 @@
// Mgmt
// Copyright (C) James Shubin and the project contributors
// Written by James Shubin <james@shubin.ca> and the project contributors
//
// This program is free software: you can redistribute it and/or modify
// it under the terms of the GNU General Public License as published by
// the Free Software Foundation, either version 3 of the License, or
// (at your option) any later version.
//
// This program is distributed in the hope that it will be useful,
// but WITHOUT ANY WARRANTY; without even the implied warranty of
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
// GNU General Public License for more details.
//
// You should have received a copy of the GNU General Public License
// along with this program. If not, see <https://www.gnu.org/licenses/>.
//
// Additional permission under GNU GPL version 3 section 7
//
// If you modify this program, or any covered work, by linking or combining it
// with embedded mcl code and modules (and that the embedded mcl code and
// modules which link with this program, contain a copy of their source code in
// the authoritative form) containing parts covered by the terms of any other
// license, the licensors of this program grant you additional permission to
// convey the resulting work. Furthermore, the licensors of this program grant
// the original author, James Shubin, additional permission to update this
// additional permission if he deems it necessary to achieve the goals of this
// additional permission.
package cli
import (
"context"
"os"
"os/signal"
"sync"
"syscall"
cliUtil "github.com/purpleidea/mgmt/cli/util"
"github.com/purpleidea/mgmt/setup"
)
// SetupArgs is the CLI parsing structure and type of the parsed result. This
// particular one contains all the common flags for the `setup` subcommand.
type SetupArgs struct {
setup.Config // embedded config (can't be a pointer) https://github.com/alexflint/go-arg/issues/240
SetupPkg *cliUtil.SetupPkgArgs `arg:"subcommand:pkg" help:"setup packages"`
SetupSvc *cliUtil.SetupSvcArgs `arg:"subcommand:svc" help:"setup services"`
SetupFirstboot *cliUtil.SetupFirstbootArgs `arg:"subcommand:firstboot" help:"setup firstboot"`
}
// Run executes the correct subcommand. It errors if there's ever an error. It
// returns true if we did activate one of the subcommands. It returns false if
// we did not. This information is used so that the top-level parser can return
// usage or help information if no subcommand activates. This particular Run is
// the run for the main `setup` subcommand. The setup command does some
// bootstrap work to help get things going.
func (obj *SetupArgs) Run(ctx context.Context, data *cliUtil.Data) (bool, error) {
ctx, cancel := context.WithCancel(ctx)
defer cancel()
var name string
var args interface{}
if cmd := obj.SetupPkg; cmd != nil {
name = cliUtil.LookupSubcommand(obj, cmd) // "pkg"
args = cmd
}
if cmd := obj.SetupSvc; cmd != nil {
name = cliUtil.LookupSubcommand(obj, cmd) // "svc"
args = cmd
}
if cmd := obj.SetupFirstboot; cmd != nil {
name = cliUtil.LookupSubcommand(obj, cmd) // "firstboot"
args = cmd
}
_ = name
Logf := func(format string, v ...interface{}) {
// Don't block this globally...
//if !data.Flags.Debug {
// return
//}
data.Flags.Logf("main: "+format, v...)
}
var api setup.API
if cmd := obj.SetupPkg; cmd != nil {
api = &setup.Pkg{
SetupPkgArgs: args.(*cliUtil.SetupPkgArgs),
Config: obj.Config,
Program: data.Program,
Version: data.Version,
Debug: data.Flags.Debug,
Logf: Logf,
}
}
if cmd := obj.SetupSvc; cmd != nil {
api = &setup.Svc{
SetupSvcArgs: args.(*cliUtil.SetupSvcArgs),
Config: obj.Config,
Program: data.Program,
Version: data.Version,
Debug: data.Flags.Debug,
Logf: Logf,
}
}
if cmd := obj.SetupFirstboot; cmd != nil {
api = &setup.Firstboot{
SetupFirstbootArgs: args.(*cliUtil.SetupFirstbootArgs),
Config: obj.Config,
Program: data.Program,
Version: data.Version,
Debug: data.Flags.Debug,
Logf: Logf,
}
}
if api == nil {
return false, nil // nothing found (display help!)
}
// We don't use these for the setup command in normal operation.
if data.Flags.Debug {
cliUtil.Hello(data.Program, data.Version, data.Flags) // say hello!
defer Logf("goodbye!")
}
// install the exit signal handler
wg := &sync.WaitGroup{}
defer wg.Wait()
exit := make(chan struct{})
defer close(exit)
wg.Add(1)
go func() {
defer cancel()
defer wg.Done()
// must have buffer for max number of signals
signals := make(chan os.Signal, 3+1) // 3 * ^C + 1 * SIGTERM
signal.Notify(signals, os.Interrupt) // catch ^C
//signal.Notify(signals, os.Kill) // catch signals
signal.Notify(signals, syscall.SIGTERM)
var count uint8
for {
select {
case sig := <-signals: // any signal will do
if sig != os.Interrupt {
data.Flags.Logf("interrupted by signal")
return
}
switch count {
case 0:
data.Flags.Logf("interrupted by ^C")
cancel()
case 1:
data.Flags.Logf("interrupted by ^C (fast pause)")
cancel()
case 2:
data.Flags.Logf("interrupted by ^C (hard interrupt)")
cancel()
}
count++
case <-exit:
return
}
}
}()
if err := api.Main(ctx); err != nil {
if data.Flags.Debug {
data.Flags.Logf("main: %+v", err)
}
return false, err
}
return true, nil
}

150
cli/tools.go Normal file
View File

@@ -0,0 +1,150 @@
// Mgmt
// Copyright (C) James Shubin and the project contributors
// Written by James Shubin <james@shubin.ca> and the project contributors
//
// This program is free software: you can redistribute it and/or modify
// it under the terms of the GNU General Public License as published by
// the Free Software Foundation, either version 3 of the License, or
// (at your option) any later version.
//
// This program is distributed in the hope that it will be useful,
// but WITHOUT ANY WARRANTY; without even the implied warranty of
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
// GNU General Public License for more details.
//
// You should have received a copy of the GNU General Public License
// along with this program. If not, see <https://www.gnu.org/licenses/>.
//
// Additional permission under GNU GPL version 3 section 7
//
// If you modify this program, or any covered work, by linking or combining it
// with embedded mcl code and modules (and that the embedded mcl code and
// modules which link with this program, contain a copy of their source code in
// the authoritative form) containing parts covered by the terms of any other
// license, the licensors of this program grant you additional permission to
// convey the resulting work. Furthermore, the licensors of this program grant
// the original author, James Shubin, additional permission to update this
// additional permission if he deems it necessary to achieve the goals of this
// additional permission.
package cli
import (
"context"
"os"
"os/signal"
"sync"
"syscall"
cliUtil "github.com/purpleidea/mgmt/cli/util"
"github.com/purpleidea/mgmt/tools"
)
// ToolsArgs is the CLI parsing structure and type of the parsed result. This
// particular one contains all the common flags for the `tools` subcommand.
type ToolsArgs struct {
tools.Config // embedded config (can't be a pointer) https://github.com/alexflint/go-arg/issues/240
ToolsGrow *cliUtil.ToolsGrowArgs `arg:"subcommand:grow" help:"tools for growing storage"`
}
// Run executes the correct subcommand. It errors if there's ever an error. It
// returns true if we did activate one of the subcommands. It returns false if
// we did not. This information is used so that the top-level parser can return
// usage or help information if no subcommand activates. This particular Run is
// the run for the main `tools` subcommand. The tools command provides some
// functionality which can be helpful with provisioning and config management.
func (obj *ToolsArgs) Run(ctx context.Context, data *cliUtil.Data) (bool, error) {
ctx, cancel := context.WithCancel(ctx)
defer cancel()
var name string
var args interface{}
if cmd := obj.ToolsGrow; cmd != nil {
name = cliUtil.LookupSubcommand(obj, cmd) // "grow"
args = cmd
}
_ = name
Logf := func(format string, v ...interface{}) {
// Don't block this globally...
//if !data.Flags.Debug {
// return
//}
data.Flags.Logf("main: "+format, v...)
}
var api tools.API
if cmd := obj.ToolsGrow; cmd != nil {
api = &tools.Grow{
ToolsGrowArgs: args.(*cliUtil.ToolsGrowArgs),
Config: obj.Config,
Program: data.Program,
Version: data.Version,
Debug: data.Flags.Debug,
Logf: Logf,
}
}
if api == nil {
return false, nil // nothing found (display help!)
}
// We don't use these for the tools command in normal operation.
if data.Flags.Debug {
cliUtil.Hello(data.Program, data.Version, data.Flags) // say hello!
defer Logf("goodbye!")
}
// install the exit signal handler
wg := &sync.WaitGroup{}
defer wg.Wait()
exit := make(chan struct{})
defer close(exit)
wg.Add(1)
go func() {
defer cancel()
defer wg.Done()
// must have buffer for max number of signals
signals := make(chan os.Signal, 3+1) // 3 * ^C + 1 * SIGTERM
signal.Notify(signals, os.Interrupt) // catch ^C
//signal.Notify(signals, os.Kill) // catch signals
signal.Notify(signals, syscall.SIGTERM)
var count uint8
for {
select {
case sig := <-signals: // any signal will do
if sig != os.Interrupt {
data.Flags.Logf("interrupted by signal")
return
}
switch count {
case 0:
data.Flags.Logf("interrupted by ^C")
cancel()
case 1:
data.Flags.Logf("interrupted by ^C (fast pause)")
cancel()
case 2:
data.Flags.Logf("interrupted by ^C (hard interrupt)")
cancel()
}
count++
case <-exit:
return
}
}
}()
if err := api.Main(ctx); err != nil {
if data.Flags.Debug {
data.Flags.Logf("main: %+v", err)
}
return false, err
}
return true, nil
}

214
cli/util/args.go Normal file
View File

@@ -0,0 +1,214 @@
// Mgmt
// Copyright (C) James Shubin and the project contributors
// Written by James Shubin <james@shubin.ca> and the project contributors
//
// This program is free software: you can redistribute it and/or modify
// it under the terms of the GNU General Public License as published by
// the Free Software Foundation, either version 3 of the License, or
// (at your option) any later version.
//
// This program is distributed in the hope that it will be useful,
// but WITHOUT ANY WARRANTY; without even the implied warranty of
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
// GNU General Public License for more details.
//
// You should have received a copy of the GNU General Public License
// along with this program. If not, see <https://www.gnu.org/licenses/>.
//
// Additional permission under GNU GPL version 3 section 7
//
// If you modify this program, or any covered work, by linking or combining it
// with embedded mcl code and modules (and that the embedded mcl code and
// modules which link with this program, contain a copy of their source code in
// the authoritative form) containing parts covered by the terms of any other
// license, the licensors of this program grant you additional permission to
// convey the resulting work. Furthermore, the licensors of this program grant
// the original author, James Shubin, additional permission to update this
// additional permission if he deems it necessary to achieve the goals of this
// additional permission.
package util
import (
"reflect"
"strings"
)
// LookupSubcommand returns the name of the subcommand in the obj, of a struct.
// This is useful for determining the name of the subcommand that was activated.
// It returns an empty string if a specific name was not found.
func LookupSubcommand(obj interface{}, st interface{}) string {
val := reflect.ValueOf(obj)
if val.Kind() == reflect.Ptr { // max one de-referencing
val = val.Elem()
}
v := reflect.ValueOf(st) // value of the struct
typ := val.Type()
for i := 0; i < typ.NumField(); i++ {
f := val.Field(i) // value of the field
if f.Interface() != v.Interface() {
continue
}
field := typ.Field(i)
alias, ok := field.Tag.Lookup("arg")
if !ok {
continue
}
// XXX: `arg` needs a split by comma first or fancier parsing
prefix := "subcommand"
split := strings.Split(alias, ":")
if len(split) != 2 || split[0] != prefix {
continue
}
return split[1] // found
}
return "" // not found
}
// EmptyArgs is the empty CLI parsing structure and type of the parsed result.
type EmptyArgs struct {
Wait bool `arg:"--wait" help:"don't use any existing (stale) deploys"`
}
// LangArgs is the lang CLI parsing structure and type of the parsed result.
type LangArgs struct {
// Input is the input mcl code or file path or any input specification.
Input string `arg:"positional,required"`
// TODO: removed (temporarily?)
//Stdin bool `arg:"--stdin" help:"use passthrough stdin"`
Download bool `arg:"--download" help:"download any missing imports"`
OnlyDownload bool `arg:"--only-download" help:"stop after downloading any missing imports"`
Update bool `arg:"--update" help:"update all dependencies to the latest versions"`
OnlyUnify bool `arg:"--only-unify" help:"stop after type unification"`
SkipUnify bool `arg:"--skip-unify" help:"skip type unification"`
UnifySolver *string `arg:"--unify-name" help:"pick a specific unification solver"`
UnifyOptimizations []string `arg:"--unify-optimizations,separate" help:"list of unification optimizations to request (experts only)"`
Depth int `arg:"--depth" default:"-1" help:"max recursion depth limit (-1 is unlimited)"`
// The default of 0 means any error is a failure by default.
Retry int `arg:"--depth" help:"max number of retries (-1 is unlimited)"`
ModulePath string `arg:"--module-path,env:MGMT_MODULE_PATH" help:"choose the modules path (absolute)"`
}
// YamlArgs is the yaml CLI parsing structure and type of the parsed result.
type YamlArgs struct {
// Input is the input yaml code or file path or any input specification.
Input string `arg:"positional,required"`
}
// PuppetArgs is the puppet CLI parsing structure and type of the parsed result.
type PuppetArgs struct {
// Input is the input puppet code or file path or just "agent".
Input string `arg:"positional,required"`
// PuppetConf is the optional path to a puppet.conf config file.
PuppetConf string `arg:"--puppet-conf" help:"full path to the puppet.conf file to use"`
}
// LangPuppetArgs is the langpuppet CLI parsing structure and type of the parsed
// result.
type LangPuppetArgs struct {
// LangInput is the input mcl code or file path or any input specification.
LangInput string `arg:"--lang,required" help:"the input parameter for the lang module"`
// PuppetInput is the input puppet code or file path or just "agent".
PuppetInput string `arg:"--puppet,required" help:"the input parameter for the puppet module"`
// copy-pasted from PuppetArgs
// PuppetConf is the optional path to a puppet.conf config file.
PuppetConf string `arg:"--puppet-conf" help:"full path to the puppet.conf file to use"`
// end PuppetArgs
// copy-pasted from LangArgs
// TODO: removed (temporarily?)
//Stdin bool `arg:"--stdin" help:"use passthrough stdin"`
Download bool `arg:"--download" help:"download any missing imports"`
OnlyDownload bool `arg:"--only-download" help:"stop after downloading any missing imports"`
Update bool `arg:"--update" help:"update all dependencies to the latest versions"`
OnlyUnify bool `arg:"--only-unify" help:"stop after type unification"`
SkipUnify bool `arg:"--skip-unify" help:"skip type unification"`
Depth int `arg:"--depth" default:"-1" help:"max recursion depth limit (-1 is unlimited)"`
// The default of 0 means any error is a failure by default.
Retry int `arg:"--depth" help:"max number of retries (-1 is unlimited)"`
ModulePath string `arg:"--module-path,env:MGMT_MODULE_PATH" help:"choose the modules path (absolute)"`
// end LangArgs
}
// SetupPkgArgs is the setup service CLI parsing structure and type of the
// parsed result.
type SetupPkgArgs struct {
Distro string `arg:"--distro" help:"build for this distro"`
Sudo bool `arg:"--sudo" help:"include sudo in the command"`
Exec bool `arg:"--exec" help:"actually run these commands"`
}
// SetupSvcArgs is the setup service CLI parsing structure and type of the
// parsed result.
type SetupSvcArgs struct {
BinaryPath string `arg:"--binary-path" help:"path to the binary"`
SSHURL string `arg:"--ssh-url" help:"transport the etcd client connection over SSH to this server"`
SSHHostKey string `arg:"--ssh-hostkey" help:"use this ssh known hosts key when connecting over SSH"`
Seeds []string `arg:"--seeds,separate,env:MGMT_SEEDS" help:"default etcd client endpoints"`
NoServer bool `arg:"--no-server" help:"do not start embedded etcd server (do not promote from client to peer)"`
Install bool `arg:"--install" help:"install the systemd mgmt service"`
Start bool `arg:"--start" help:"start the mgmt service"`
Enable bool `arg:"--enable" help:"enable the mgmt service"`
}
// SetupFirstbootArgs is the setup service CLI parsing structure and type of the
// parsed result.
type SetupFirstbootArgs struct {
BinaryPath string `arg:"--binary-path" help:"path to the binary"`
Mkdir bool `arg:"--mkdir" help:"make the necessary firstboot dirs"`
Install bool `arg:"--install" help:"install the systemd firstboot service"`
Start bool `arg:"--start" help:"start the firstboot service (typically not used)"`
Enable bool `arg:"--enable" help:"enable the firstboot service"`
FirstbootStartArgs // Include these options if we want to specify them.
}
// FirstbootStartArgs is the firstboot service CLI parsing structure and type of
// the parsed result.
type FirstbootStartArgs struct {
LockFilePath string `arg:"--lock-file-path" help:"path to the lock file"`
ScriptsDir string `arg:"--scripts-dir" help:"path to the scripts dir"`
DoneDir string `arg:"--done-dir" help:"dir to move done scripts to"`
LoggingDir string `arg:"--logging-dir" help:"directory to store logs in"`
}
// DocsGenerateArgs is the docgen utility CLI parsing structure and type of the
// parsed result.
type DocsGenerateArgs struct {
Output string `arg:"--output" help:"output path to write to"`
RootDir string `arg:"--root-dir" help:"path to mgmt source dir"`
NoResources bool `arg:"--no-resources" help:"skip resource doc generation"`
NoFunctions bool `arg:"--no-functions" help:"skip function doc generation"`
}
// ToolsGrowArgs is the util tool CLI parsing structure and type of the parsed
// result.
type ToolsGrowArgs struct {
Mount string `arg:"--mount,required" help:"root mount point to start with"`
Exec bool `arg:"--exec" help:"actually run these commands"`
Done string `arg:"--done" help:"create this file when done, skip if it exists"`
}

47
cli/util/hello.go Normal file
View File

@@ -0,0 +1,47 @@
// Mgmt
// Copyright (C) James Shubin and the project contributors
// Written by James Shubin <james@shubin.ca> and the project contributors
//
// This program is free software: you can redistribute it and/or modify
// it under the terms of the GNU General Public License as published by
// the Free Software Foundation, either version 3 of the License, or
// (at your option) any later version.
//
// This program is distributed in the hope that it will be useful,
// but WITHOUT ANY WARRANTY; without even the implied warranty of
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
// GNU General Public License for more details.
//
// You should have received a copy of the GNU General Public License
// along with this program. If not, see <https://www.gnu.org/licenses/>.
//
// Additional permission under GNU GPL version 3 section 7
//
// If you modify this program, or any covered work, by linking or combining it
// with embedded mcl code and modules (and that the embedded mcl code and
// modules which link with this program, contain a copy of their source code in
// the authoritative form) containing parts covered by the terms of any other
// license, the licensors of this program grant you additional permission to
// convey the resulting work. Furthermore, the licensors of this program grant
// the original author, James Shubin, additional permission to update this
// additional permission if he deems it necessary to achieve the goals of this
// additional permission.
package util
import (
"fmt"
"time"
)
// Hello is a simple helper function to print a hello message and time.
func Hello(program, version string, flags Flags) {
var start = time.Now().UnixNano()
if program == "" {
program = "<unknown>"
}
fmt.Println(fmt.Sprintf("This is: %s, version: %s", program, version))
fmt.Println("Copyright (C) James Shubin and the project contributors")
fmt.Println("Written by James Shubin <james@shubin.ca> and the project contributors")
flags.Logf("main: start: %v", start)
}

98
cli/util/util.go Normal file
View File

@@ -0,0 +1,98 @@
// Mgmt
// Copyright (C) James Shubin and the project contributors
// Written by James Shubin <james@shubin.ca> and the project contributors
//
// This program is free software: you can redistribute it and/or modify
// it under the terms of the GNU General Public License as published by
// the Free Software Foundation, either version 3 of the License, or
// (at your option) any later version.
//
// This program is distributed in the hope that it will be useful,
// but WITHOUT ANY WARRANTY; without even the implied warranty of
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
// GNU General Public License for more details.
//
// You should have received a copy of the GNU General Public License
// along with this program. If not, see <https://www.gnu.org/licenses/>.
//
// Additional permission under GNU GPL version 3 section 7
//
// If you modify this program, or any covered work, by linking or combining it
// with embedded mcl code and modules (and that the embedded mcl code and
// modules which link with this program, contain a copy of their source code in
// the authoritative form) containing parts covered by the terms of any other
// license, the licensors of this program grant you additional permission to
// convey the resulting work. Furthermore, the licensors of this program grant
// the original author, James Shubin, additional permission to update this
// additional permission if he deems it necessary to achieve the goals of this
// additional permission.
// Package util has some CLI related utility code.
package util
import (
"log"
"os"
"strings"
"github.com/purpleidea/mgmt/util/errwrap"
)
// Error is a constant error type that implements error.
type Error string
// Error fulfills the error interface of this type.
func (e Error) Error() string { return string(e) }
const (
// MissingEquals means we probably hit the parsing bug.
// XXX: see: https://github.com/alexflint/go-arg/issues/239
MissingEquals = Error("missing equals sign for list element")
)
// CliParseError returns a consistent error if we have a CLI parsing issue.
func CliParseError(err error) error {
return errwrap.Wrapf(err, "cli parse error")
}
// Flags are some constant flags which are used throughout the program.
type Flags struct {
Debug bool // add additional log messages
Logf func(format string, v ...interface{})
}
// Data is a struct of values that we usually pass to the main CLI function.
type Data struct {
Program string
Version string
Copying string
Tagline string
Flags Flags
Args []string // os.Args usually
}
// SafeProgram returns the correct program string when given a buggy variant.
func SafeProgram(program string) string {
// FIXME: in sub commands, the cli package appends a space and the sub
// command name at the end. hack around this by only using the first bit
// see: https://github.com/urfave/cli/issues/783 for more details...
split := strings.Split(program, " ")
program = split[0]
//if program == "" {
// program = "<unknown>"
//}
return program
}
// LogSetup changes some of the core logger package settings.
func LogSetup(debug bool) {
// TODO: Move these log package initialization steps to the top main.go?
logFlags := log.LstdFlags
if debug {
logFlags = logFlags + log.Lshortfile
}
logFlags = logFlags - log.Ldate // remove the date for now
log.SetFlags(logFlags)
log.SetOutput(os.Stderr)
}

229
config.go
View File

@@ -1,229 +0,0 @@
// Mgmt
// Copyright (C) 2013-2016+ James Shubin and the project contributors
// Written by James Shubin <james@shubin.ca> and the project contributors
//
// This program is free software: you can redistribute it and/or modify
// it under the terms of the GNU Affero General Public License as published by
// the Free Software Foundation, either version 3 of the License, or
// (at your option) any later version.
//
// This program is distributed in the hope that it will be useful,
// but WITHOUT ANY WARRANTY; without even the implied warranty of
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
// GNU Affero General Public License for more details.
//
// You should have received a copy of the GNU Affero General Public License
// along with this program. If not, see <http://www.gnu.org/licenses/>.
package main
import (
"errors"
"gopkg.in/yaml.v2"
"io/ioutil"
"log"
"strings"
)
type collectorTypeConfig struct {
Type string `yaml:"type"`
Pattern string `yaml:"pattern"` // XXX: Not Implemented
}
type vertexConfig struct {
Type string `yaml:"type"`
Name string `yaml:"name"`
}
type edgeConfig struct {
Name string `yaml:"name"`
From vertexConfig `yaml:"from"`
To vertexConfig `yaml:"to"`
}
type GraphConfig struct {
Graph string `yaml:"graph"`
Types struct {
Noop []NoopType `yaml:"noop"`
File []FileType `yaml:"file"`
Service []ServiceType `yaml:"service"`
Exec []ExecType `yaml:"exec"`
} `yaml:"types"`
Collector []collectorTypeConfig `yaml:"collect"`
Edges []edgeConfig `yaml:"edges"`
Comment string `yaml:"comment"`
}
func (c *GraphConfig) Parse(data []byte) error {
if err := yaml.Unmarshal(data, c); err != nil {
return err
}
if c.Graph == "" {
return errors.New("Graph config: invalid `graph`")
}
return nil
}
func ParseConfigFromFile(filename string) *GraphConfig {
data, err := ioutil.ReadFile(filename)
if err != nil {
log.Printf("Error: Config: ParseConfigFromFile: File: %v", err)
return nil
}
var config GraphConfig
if err := config.Parse(data); err != nil {
log.Printf("Error: Config: ParseConfigFromFile: Parse: %v", err)
return nil
}
return &config
}
// XXX: we need to fix this function so that it either fails without modifying
// the graph, passes successfully and modifies it, or basically panics i guess
// this way an invalid compilation can leave the old graph running, and we we
// don't modify a partial graph. so we really need to validate, and then perform
// whatever actions are necessary
// finding some way to do this on a copy of the graph, and then do a graph diff
// and merge the new data into the old graph would be more appropriate, in
// particular if we can ensure the graph merge can't fail. As for the putting
// of stuff into etcd, we should probably store the operations to complete in
// the new graph, and keep retrying until it succeeds, thus blocking any new
// etcd operations until that time.
func UpdateGraphFromConfig(config *GraphConfig, hostname string, g *Graph, etcdO *EtcdWObject) bool {
var NoopMap = make(map[string]*Vertex)
var FileMap = make(map[string]*Vertex)
var ServiceMap = make(map[string]*Vertex)
var ExecMap = make(map[string]*Vertex)
var lookup = make(map[string]map[string]*Vertex)
lookup["noop"] = NoopMap
lookup["file"] = FileMap
lookup["service"] = ServiceMap
lookup["exec"] = ExecMap
//log.Printf("%+v", config) // debug
g.SetName(config.Graph) // set graph name
var keep []*Vertex // list of vertex which are the same in new graph
for _, t := range config.Types.Noop {
obj := NewNoopType(t.Name)
v := g.GetVertexMatch(obj)
if v == nil { // no match found
v = NewVertex(obj)
g.AddVertex(v) // call standalone in case not part of an edge
}
NoopMap[obj.Name] = v // used for constructing edges
keep = append(keep, v) // append
}
for _, t := range config.Types.File {
// XXX: should we export based on a @@ prefix, or a metaparam
// like exported => true || exported => (host pattern)||(other pattern?)
if strings.HasPrefix(t.Name, "@@") { // exported resource
// add to etcd storage...
t.Name = t.Name[2:] //slice off @@
if !etcdO.EtcdPut(hostname, t.Name, "file", t) {
log.Printf("Problem exporting file resource %v.", t.Name)
continue
}
} else {
obj := NewFileType(t.Name, t.Path, t.Dirname, t.Basename, t.Content, t.State)
v := g.GetVertexMatch(obj)
if v == nil { // no match found
v = NewVertex(obj)
g.AddVertex(v) // call standalone in case not part of an edge
}
FileMap[obj.Name] = v // used for constructing edges
keep = append(keep, v) // append
}
}
for _, t := range config.Types.Service {
obj := NewServiceType(t.Name, t.State, t.Startup)
v := g.GetVertexMatch(obj)
if v == nil { // no match found
v = NewVertex(obj)
g.AddVertex(v) // call standalone in case not part of an edge
}
ServiceMap[obj.Name] = v // used for constructing edges
keep = append(keep, v) // append
}
for _, t := range config.Types.Exec {
obj := NewExecType(t.Name, t.Cmd, t.Shell, t.Timeout, t.WatchCmd, t.WatchShell, t.IfCmd, t.IfShell, t.PollInt, t.State)
v := g.GetVertexMatch(obj)
if v == nil { // no match found
v = NewVertex(obj)
g.AddVertex(v) // call standalone in case not part of an edge
}
ExecMap[obj.Name] = v // used for constructing edges
keep = append(keep, v) // append
}
// lookup from etcd graph
// do all the graph look ups in one single step, so that if the etcd
// database changes, we don't have a partial state of affairs...
nodes, ok := etcdO.EtcdGet()
if ok {
for _, t := range config.Collector {
// XXX: use t.Type and optionally t.Pattern to collect from etcd storage
log.Printf("Collect: %v; Pattern: %v", t.Type, t.Pattern)
for _, x := range etcdO.EtcdGetProcess(nodes, "file") {
var obj *FileType
if B64ToObj(x, &obj) != true {
log.Printf("Collect: File: %v not collected!", x)
continue
}
if t.Pattern != "" { // XXX: currently the pattern for files can only override the Dirname variable :P
obj.Dirname = t.Pattern
}
log.Printf("Collect: File: %v collected!", obj.GetName())
// XXX: similar to file add code:
v := g.GetVertexMatch(obj)
if v == nil { // no match found
obj.Init() // initialize go channels or things won't work!!!
v = NewVertex(obj)
g.AddVertex(v) // call standalone in case not part of an edge
}
FileMap[obj.GetName()] = v // used for constructing edges
keep = append(keep, v) // append
}
}
}
// get rid of any vertices we shouldn't "keep" (that aren't in new graph)
for _, v := range g.GetVertices() {
if !HasVertex(v, keep) {
// wait for exit before starting new graph!
v.Type.SendEvent(eventExit, true, false)
g.DeleteVertex(v)
}
}
for _, e := range config.Edges {
if _, ok := lookup[e.From.Type]; !ok {
return false
}
if _, ok := lookup[e.To.Type]; !ok {
return false
}
if _, ok := lookup[e.From.Type][e.From.Name]; !ok {
return false
}
if _, ok := lookup[e.To.Type][e.To.Name]; !ok {
return false
}
g.AddEdge(lookup[e.From.Type][e.From.Name], lookup[e.To.Type][e.To.Name], NewEdge(e.Name))
}
return true
}

View File

@@ -1,155 +0,0 @@
// Mgmt
// Copyright (C) 2013-2016+ James Shubin and the project contributors
// Written by James Shubin <james@shubin.ca> and the project contributors
//
// This program is free software: you can redistribute it and/or modify
// it under the terms of the GNU Affero General Public License as published by
// the Free Software Foundation, either version 3 of the License, or
// (at your option) any later version.
//
// This program is distributed in the hope that it will be useful,
// but WITHOUT ANY WARRANTY; without even the implied warranty of
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
// GNU Affero General Public License for more details.
//
// You should have received a copy of the GNU Affero General Public License
// along with this program. If not, see <http://www.gnu.org/licenses/>.
package main
import (
"gopkg.in/fsnotify.v1"
//"github.com/go-fsnotify/fsnotify" // git master of "gopkg.in/fsnotify.v1"
"log"
"math"
"path"
"strings"
"syscall"
)
// XXX: it would be great if we could reuse code between this and the file type
// XXX: patch this to submit it as part of go-fsnotify if they're interested...
func ConfigWatch(file string) chan bool {
ch := make(chan bool)
go func() {
var safename = path.Clean(file) // no trailing slash
watcher, err := fsnotify.NewWatcher()
if err != nil {
log.Fatal(err)
}
defer watcher.Close()
patharray := PathSplit(safename) // tokenize the path
var index = len(patharray) // starting index
var current string // current "watcher" location
var deltaDepth int // depth delta between watcher and event
var send = false // send event?
for {
current = strings.Join(patharray[0:index], "/")
if current == "" { // the empty string top is the root dir ("/")
current = "/"
}
log.Printf("Watching: %v", current) // attempting to watch...
// initialize in the loop so that we can reset on rm-ed handles
err = watcher.Add(current)
if err != nil {
if err == syscall.ENOENT {
index-- // usually not found, move up one dir
} else if err == syscall.ENOSPC {
// XXX: occasionally: no space left on device,
// XXX: probably due to lack of inotify watches
log.Printf("Lack of watches for config(%v) error: %+v", file, err.Error) // 0x408da0
log.Fatal(err)
} else {
log.Printf("Unknown config(%v) error:", file)
log.Fatal(err)
}
index = int(math.Max(1, float64(index)))
continue
}
select {
case event := <-watcher.Events:
// the deeper you go, the bigger the deltaDepth is...
// this is the difference between what we're watching,
// and the event... doesn't mean we can't watch deeper
if current == event.Name {
deltaDepth = 0 // i was watching what i was looking for
} else if HasPathPrefix(event.Name, current) {
deltaDepth = len(PathSplit(current)) - len(PathSplit(event.Name)) // -1 or less
} else if HasPathPrefix(current, event.Name) {
deltaDepth = len(PathSplit(event.Name)) - len(PathSplit(current)) // +1 or more
} else {
// TODO different watchers get each others events!
// https://github.com/go-fsnotify/fsnotify/issues/95
// this happened with two values such as:
// event.Name: /tmp/mgmt/f3 and current: /tmp/mgmt/f2
continue
}
//log.Printf("The delta depth is: %v", deltaDepth)
// if we have what we wanted, awesome, send an event...
if event.Name == safename {
//log.Println("Event!")
send = true
// file removed, move the watch upwards
if deltaDepth >= 0 && (event.Op&fsnotify.Remove == fsnotify.Remove) {
//log.Println("Removal!")
watcher.Remove(current)
index--
}
// we must be a parent watcher, so descend in
if deltaDepth < 0 {
watcher.Remove(current)
index++
}
// if safename starts with event.Name, we're above, and no event should be sent
} else if HasPathPrefix(safename, event.Name) {
//log.Println("Above!")
if deltaDepth >= 0 && (event.Op&fsnotify.Remove == fsnotify.Remove) {
log.Println("Removal!")
watcher.Remove(current)
index--
}
if deltaDepth < 0 {
log.Println("Parent!")
if PathPrefixDelta(safename, event.Name) == 1 { // we're the parent dir
//send = true
}
watcher.Remove(current)
index++
}
// if event.Name startswith safename, send event, we're already deeper
} else if HasPathPrefix(event.Name, safename) {
//log.Println("Event2!")
//send = true
}
case err := <-watcher.Errors:
log.Println("error:", err)
log.Fatal(err)
}
// do our event sending all together to avoid duplicate msgs
if send {
send = false
ch <- true
}
}
//close(ch)
}()
return ch
}

530
converger/converger.go Normal file
View File

@@ -0,0 +1,530 @@
// Mgmt
// Copyright (C) James Shubin and the project contributors
// Written by James Shubin <james@shubin.ca> and the project contributors
//
// This program is free software: you can redistribute it and/or modify
// it under the terms of the GNU General Public License as published by
// the Free Software Foundation, either version 3 of the License, or
// (at your option) any later version.
//
// This program is distributed in the hope that it will be useful,
// but WITHOUT ANY WARRANTY; without even the implied warranty of
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
// GNU General Public License for more details.
//
// You should have received a copy of the GNU General Public License
// along with this program. If not, see <https://www.gnu.org/licenses/>.
//
// Additional permission under GNU GPL version 3 section 7
//
// If you modify this program, or any covered work, by linking or combining it
// with embedded mcl code and modules (and that the embedded mcl code and
// modules which link with this program, contain a copy of their source code in
// the authoritative form) containing parts covered by the terms of any other
// license, the licensors of this program grant you additional permission to
// convey the resulting work. Furthermore, the licensors of this program grant
// the original author, James Shubin, additional permission to update this
// additional permission if he deems it necessary to achieve the goals of this
// additional permission.
// Package converger is a facility for reporting the converged state.
package converger
import (
"fmt"
"sort"
"sync"
"sync/atomic"
"time"
"github.com/purpleidea/mgmt/util"
"github.com/purpleidea/mgmt/util/errwrap"
)
// New builds a new converger coordinator.
func New(timeout int) *Coordinator {
return &Coordinator{
timeout: timeout,
mutex: &sync.RWMutex{},
//lastid: 0,
status: make(map[*UID]struct{}),
//converged: false, // initial state
pokeChan: make(chan struct{}, 1), // must be buffered
readyChan: make(chan struct{}), // ready signal
//paused: false, // starts off as started
pauseSignal: make(chan struct{}),
//resumeSignal: make(chan struct{}), // happens on pause
//pausedAck: util.NewEasyAck(), // happens on pause
sendSignal: make(chan bool),
stateFns: make(map[string]func(bool) error),
smutex: &sync.RWMutex{},
closeChan: make(chan struct{}),
wg: &sync.WaitGroup{},
}
}
// Coordinator is the central converger engine.
type Coordinator struct {
// timeout must be zero (instant) or greater seconds to run. If it's -1
// then this is disabled, and we never run stateFns.
timeout int
// mutex is used for controlling access to status and lastid.
mutex *sync.RWMutex
// lastid contains the last uid we used for registration.
//lastid uint64
// status contains a reference to each active UID.
status map[*UID]struct{}
// converged stores the last convergence state. When this changes, we
// run the stateFns.
converged bool
// pokeChan receives a message every time we might need to re-calculate.
pokeChan chan struct{}
// readyChan closes to notify any interested parties that the main loop
// is running.
readyChan chan struct{}
// paused represents if this coordinator is paused or not.
paused bool
// pauseSignal closes to request a pause of this coordinator.
pauseSignal chan struct{}
// resumeSignal closes to request a resume of this coordinator.
resumeSignal chan struct{}
// pausedAck is used to send an ack message saying that we've paused.
pausedAck *util.EasyAck
sendSignal chan bool // send pause (false) or resume (true)
// stateFns run on converged state changes.
stateFns map[string]func(bool) error
// smutex is used for controlling access to the stateFns map.
smutex *sync.RWMutex
// closeChan closes when we've been requested to shutdown.
closeChan chan struct{}
// wg waits for everything to finish.
wg *sync.WaitGroup
}
// Register creates a new UID which can be used to report converged state. You
// must Unregister each UID before Shutdown will be able to finish running.
func (obj *Coordinator) Register() *UID {
obj.wg.Add(1) // additional tracking for each UID
obj.mutex.Lock()
defer obj.mutex.Unlock()
//obj.lastid++
uid := &UID{
timeout: obj.timeout, // copy the timeout here
//id: obj.lastid,
//name: fmt.Sprintf("%d", obj.lastid), // some default
isConverged: &atomic.Bool{},
poke: obj.poke,
// timer
mutex: &sync.Mutex{},
timer: nil,
running: false,
wg: &sync.WaitGroup{},
}
uid.unregister = func() { obj.Unregister(uid) } // add unregister func
obj.status[uid] = struct{}{} // TODO: add converged state here?
return uid
}
// Unregister removes the UID from the converger coordinator. If you supply an
// invalid or unregistered uid to this function, it will panic. An unregistered
// UID is no longer part of the convergence checking.
func (obj *Coordinator) Unregister(uid *UID) {
defer obj.wg.Done() // additional tracking for each UID
obj.mutex.Lock()
defer obj.mutex.Unlock()
if _, exists := obj.status[uid]; !exists {
panic("uid is not registered")
}
uid.StopTimer() // ignore any errors
delete(obj.status, uid)
}
// Run starts the main loop for the converger coordinator. It is commonly run
// from a go routine. It blocks until the Shutdown method is run to close it.
// NOTE: when we have very short timeouts, if we start before all the resources
// have joined the map, then it might appear as if we converged before we did!
func (obj *Coordinator) Run(startPaused bool) {
obj.wg.Add(1)
wg := &sync.WaitGroup{} // needed for the startPaused
defer wg.Wait() // don't leave any leftover go routines running
if startPaused {
wg.Add(1)
go func() {
defer wg.Done()
obj.Pause() // ignore any errors
close(obj.readyChan)
}()
} else {
close(obj.readyChan) // we must wait till the wg.Add(1) has happened...
}
defer obj.wg.Done()
for {
// pause if one was requested...
select {
//case <-obj.pauseSignal: // channel closes
// obj.pausedAck.Ack() // send ack
// // we are paused now, and waiting for resume or exit...
// select {
// case <-obj.resumeSignal: // channel closes # XXX: RACE READ
// // resumed!
//
// case <-obj.closeChan: // we can always escape
// return
// }
case b, _ := <-obj.sendSignal:
if b { // resume
panic("unexpected resume") // TODO: continue instead?
}
// paused
obj.pausedAck.Ack() // send ack
// we are paused now, and waiting for resume or exit...
select {
case b, _ := <-obj.sendSignal:
if !b { // pause
panic("unexpected pause") // TODO: continue instead?
}
// resumed!
case <-obj.closeChan: // we can always escape
return
}
case _, ok := <-obj.pokeChan: // we got an event (re-calculate)
if !ok {
return
}
if err := obj.test(); err != nil {
// FIXME: what to do on error ?
}
case <-obj.closeChan: // we can always escape
return
}
}
}
// Ready blocks until the Run loop has started up. This is useful so that we
// don't run Shutdown before we've even started up properly.
func (obj *Coordinator) Ready() {
select {
case <-obj.readyChan:
}
}
// Shutdown sends a signal to the Run loop that it should exit. This blocks
// until it does.
func (obj *Coordinator) Shutdown() {
close(obj.closeChan)
obj.wg.Wait()
close(obj.pokeChan) // free memory?
}
// Pause pauses the coordinator. It should not be called on an already paused
// coordinator. It will block until the coordinator pauses with an
// acknowledgment, or until an exit is requested. If the latter happens it will
// error. It is NOT thread-safe with the Resume() method so only call either one
// at a time.
func (obj *Coordinator) Pause() error {
if obj.paused {
return fmt.Errorf("already paused")
}
obj.pausedAck = util.NewEasyAck()
//obj.resumeSignal = make(chan struct{}) // build the resume signal XXX: RACE WRITE
//close(obj.pauseSignal)
select {
case obj.sendSignal <- false:
case <-obj.closeChan:
return fmt.Errorf("closing")
}
// wait for ack (or exit signal)
select {
case <-obj.pausedAck.Wait(): // we got it!
// we're paused
case <-obj.closeChan:
return fmt.Errorf("closing")
}
obj.paused = true
return nil
}
// Resume unpauses the coordinator. It can be safely called on a brand-new
// coordinator that has just started running without incident. It is NOT
// thread-safe with the Pause() method, so only call either one at a time.
func (obj *Coordinator) Resume() {
// TODO: do we need a mutex around Resume?
if !obj.paused { // no need to unpause brand-new resources
return
}
//obj.pauseSignal = make(chan struct{}) // rebuild for next pause
//close(obj.resumeSignal)
select {
case obj.sendSignal <- true:
case <-obj.closeChan:
return
}
obj.poke() // unblock and notice the resume if necessary
obj.paused = false
// no need to wait for it to resume
//return // implied
}
// poke sends a message to the coordinator telling it that it should re-evaluate
// whether we're converged or not. This does not block. Do not run this in a
// goroutine. It must not be called after Shutdown has been called.
func (obj *Coordinator) poke() {
// redundant
//if len(obj.pokeChan) > 0 {
// return
//}
select {
case obj.pokeChan <- struct{}{}:
default: // if chan is now full because more than one poke happened...
}
}
// IsConverged returns true if *every* registered uid has converged. If there
// are no registered UID's, then this will return true.
func (obj *Coordinator) IsConverged() bool {
for _, v := range obj.Status() {
if !v { // everyone must be converged for this to be true
return false
}
}
return true
}
// test evaluates whether we're converged or not and runs the state change. It
// is NOT thread-safe.
func (obj *Coordinator) test() error {
// TODO: add these checks elsewhere to prevent anything from running?
if obj.timeout < 0 {
return nil // nothing to do (only run if timeout is valid)
}
converged := obj.IsConverged()
defer func() {
obj.converged = converged // set this only at the end...
}()
if !converged {
if !obj.converged { // were we previously also not converged?
return nil // nothing to do
}
// we're doing a state change
// call the arbitrary functions (takes a read lock!)
return obj.runStateFns(false)
}
// we have converged!
if obj.converged { // were we previously also converged?
return nil // nothing to do
}
// call the arbitrary functions (takes a read lock!)
return obj.runStateFns(true)
}
// runStateFns runs the list of stored state functions.
func (obj *Coordinator) runStateFns(converged bool) error {
obj.smutex.RLock()
defer obj.smutex.RUnlock()
var keys []string
for k := range obj.stateFns {
keys = append(keys, k)
}
sort.Strings(keys)
var err error
for _, name := range keys { // run in deterministic order
fn := obj.stateFns[name]
// call an arbitrary function
e := fn(converged)
err = errwrap.Append(err, e) // list of errors
}
return err
}
// AddStateFn adds a state function to be run on change of converged state.
func (obj *Coordinator) AddStateFn(name string, stateFn func(bool) error) error {
obj.smutex.Lock()
defer obj.smutex.Unlock()
if _, exists := obj.stateFns[name]; exists {
return fmt.Errorf("a stateFn with that name already exists")
}
obj.stateFns[name] = stateFn
return nil
}
// RemoveStateFn removes a state function from running on change of converged
// state.
func (obj *Coordinator) RemoveStateFn(name string) error {
obj.smutex.Lock()
defer obj.smutex.Unlock()
if _, exists := obj.stateFns[name]; !exists {
return fmt.Errorf("a stateFn with that name doesn't exist")
}
delete(obj.stateFns, name)
return nil
}
// Status returns a map of the converged status of each UID.
func (obj *Coordinator) Status() map[*UID]bool {
status := make(map[*UID]bool)
obj.mutex.RLock() // take a read lock
defer obj.mutex.RUnlock()
for k := range obj.status {
status[k] = k.IsConverged()
}
return status
}
// Timeout returns the timeout in seconds that converger was created with. This
// is useful to avoid passing in the timeout value separately when you're
// already passing in the Coordinator struct.
func (obj *Coordinator) Timeout() int {
return obj.timeout
}
// UID represents one of the probes for the converger coordinator. It is created
// by calling the Register method of the Coordinator struct. It should be freed
// after use with Unregister.
type UID struct {
// timeout is a copy of the main timeout. It could eventually be used
// for per-UID timeouts too.
timeout int
// isConverged stores the convergence state of this particular UID.
isConverged *atomic.Bool
// poke stores a reference to the main poke function.
poke func()
// unregister stores a reference to the unregister function.
unregister func()
// timer
mutex *sync.Mutex
timer chan struct{}
running bool // is the timer running?
wg *sync.WaitGroup
}
// Unregister removes this UID from the converger coordinator. An unregistered
// UID is no longer part of the convergence checking.
func (obj *UID) Unregister() {
obj.unregister()
}
// IsConverged reports whether this UID is converged or not.
func (obj *UID) IsConverged() bool {
return obj.isConverged.Load()
}
// SetConverged sets the convergence state of this UID. This is used by the
// running timer if one is started. The timer will overwrite any value set by
// this method.
func (obj *UID) SetConverged(isConverged bool) {
obj.isConverged.Store(isConverged)
obj.poke() // notify of change
}
// ConvergedTimer adds a timeout to a select call and blocks until then.
// TODO: this means we could eventually have per resource converged timeouts
func (obj *UID) ConvergedTimer() <-chan time.Time {
// be clever: if i'm already converged, this timeout should block which
// avoids unnecessary new signals being sent! this avoids fast loops if
// we have a low timeout, or in particular a timeout == 0
if obj.IsConverged() {
// blocks the case statement in select forever!
return util.TimeAfterOrBlock(-1)
}
return util.TimeAfterOrBlock(int(obj.timeout))
}
// StartTimer runs a timer that sets us as converged on timeout. It also returns
// a handle to the StopTimer function which should be run before exit.
func (obj *UID) StartTimer() (func() error, error) {
obj.mutex.Lock()
defer obj.mutex.Unlock()
if obj.running {
return obj.StopTimer, fmt.Errorf("timer already started")
}
obj.timer = make(chan struct{})
obj.running = true
obj.wg.Add(1)
go func() {
defer obj.wg.Done()
for {
select {
case _, ok := <-obj.timer: // reset signal channel
if !ok {
return
}
obj.SetConverged(false)
case <-obj.ConvergedTimer():
obj.SetConverged(true) // converged!
select {
case _, ok := <-obj.timer: // reset signal channel
if !ok {
return
}
}
}
}
}()
return obj.StopTimer, nil
}
// ResetTimer resets the timer to zero.
func (obj *UID) ResetTimer() error {
obj.mutex.Lock()
defer obj.mutex.Unlock()
if obj.running {
obj.timer <- struct{}{} // send the reset message
return nil
}
return fmt.Errorf("timer hasn't been started")
}
// StopTimer stops the running timer.
func (obj *UID) StopTimer() error {
obj.mutex.Lock()
defer obj.mutex.Unlock()
if !obj.running {
return fmt.Errorf("timer isn't running")
}
close(obj.timer)
obj.wg.Wait()
obj.running = false
return nil
}

View File

@@ -0,0 +1,43 @@
// Mgmt
// Copyright (C) James Shubin and the project contributors
// Written by James Shubin <james@shubin.ca> and the project contributors
//
// This program is free software: you can redistribute it and/or modify
// it under the terms of the GNU General Public License as published by
// the Free Software Foundation, either version 3 of the License, or
// (at your option) any later version.
//
// This program is distributed in the hope that it will be useful,
// but WITHOUT ANY WARRANTY; without even the implied warranty of
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
// GNU General Public License for more details.
//
// You should have received a copy of the GNU General Public License
// along with this program. If not, see <https://www.gnu.org/licenses/>.
//
// Additional permission under GNU GPL version 3 section 7
//
// If you modify this program, or any covered work, by linking or combining it
// with embedded mcl code and modules (and that the embedded mcl code and
// modules which link with this program, contain a copy of their source code in
// the authoritative form) containing parts covered by the terms of any other
// license, the licensors of this program grant you additional permission to
// convey the resulting work. Furthermore, the licensors of this program grant
// the original author, James Shubin, additional permission to update this
// additional permission if he deems it necessary to achieve the goals of this
// additional permission.
//go:build !root
package converger
import (
"testing"
)
func TestBufferedChan1(t *testing.T) {
ch := make(chan bool, 1)
ch <- true
close(ch) // closing a channel that's not empty should not block
// must be able to exit without blocking anywhere
}

7
debian/.gitignore vendored Normal file
View File

@@ -0,0 +1,7 @@
*.debhelper.log
*debhelper
changelog
debhelper-build-stamp
files
mgmt.substvars
mgmt/*

1
debian/compat vendored Normal file
View File

@@ -0,0 +1 @@
9

17
debian/control vendored Normal file
View File

@@ -0,0 +1,17 @@
Source: mgmt
Maintainer: Johan Bloemberg (aequitas) <mgmt@ijohan.nl>
Build-Depends:
debhelper,
devscripts,
dh-golang,
dh-systemd,
golang-go,
Package: mgmt
Architecture: any
Depends: ${shlibs:Depends}, ${misc:Depends}, packagekit
Suggests: graphviz
Description: mgmt: next generation config management!
The mgmt tool is a next generation config management solution. It's
ready for production, and we hope you try out the future soon. Get
involved today!

33
debian/copyright vendored Normal file
View File

@@ -0,0 +1,33 @@
Format: http://www.debian.org/doc/packaging-manuals/copyright-format/1.0/
Upstream-Name: mgmt
Source: <https://github.com/purpleidea/mgmt>
Files: *
Copyright: Copyright (C) James Shubin and the project contributors
License: GPL-3.0
License: GPL-3.0
This program is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation, either version 3 of the License, or
(at your option) any later version.
This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU General Public License for more details.
You should have received a copy of the GNU General Public License
along with this program. If not, see <https://www.gnu.org/licenses/>.
Additional permission under GNU GPL version 3 section 7
If you modify this program, or any covered work, by linking or combining it
with embedded mcl code and modules (and that the embedded mcl code and
modules which link with this program, contain a copy of their source code in
the authoritative form) containing parts covered by the terms of any other
license, the licensors of this program grant you additional permission to
convey the resulting work. Furthermore, the licensors of this program grant
the original author, James Shubin, additional permission to update this
additional permission if he deems it necessary to achieve the goals of this
additional permission.

11
debian/mgmt.docs vendored Normal file
View File

@@ -0,0 +1,11 @@
AUTHORS
COPYING
COPYRIGHT
README.md
THANKS
TODO.md
docs
examples
misc/bashrc.sh
misc/delta-cpu.sh
misc/mgmt.service

2
debian/mgmt.install vendored Normal file
View File

@@ -0,0 +1,2 @@
mgmt usr/bin
misc/mgmt.service /lib/systemd/system

15
debian/rules vendored Executable file
View File

@@ -0,0 +1,15 @@
#!/usr/bin/make -f
export DH_OPTIONS
export DH_GOPKG := mgmt
export DH_GOLANG_INSTALL_ALL := 1
unexport GOROOT
override_dh_auto_build:
make build
override_dh_auto_test:
@echo "Tests are disabled for now"
%:
dh $@ --with=systemd

31
doc.go Normal file
View File

@@ -0,0 +1,31 @@
// Mgmt
// Copyright (C) James Shubin and the project contributors
// Written by James Shubin <james@shubin.ca> and the project contributors
//
// This program is free software: you can redistribute it and/or modify
// it under the terms of the GNU General Public License as published by
// the Free Software Foundation, either version 3 of the License, or
// (at your option) any later version.
//
// This program is distributed in the hope that it will be useful,
// but WITHOUT ANY WARRANTY; without even the implied warranty of
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
// GNU General Public License for more details.
//
// You should have received a copy of the GNU General Public License
// along with this program. If not, see <https://www.gnu.org/licenses/>.
//
// Additional permission under GNU GPL version 3 section 7
//
// If you modify this program, or any covered work, by linking or combining it
// with embedded mcl code and modules (and that the embedded mcl code and
// modules which link with this program, contain a copy of their source code in
// the authoritative form) containing parts covered by the terms of any other
// license, the licensors of this program grant you additional permission to
// convey the resulting work. Furthermore, the licensors of this program grant
// the original author, James Shubin, additional permission to update this
// additional permission if he deems it necessary to achieve the goals of this
// additional permission.
// Package main provides the main entrypoint for using the `mgmt` software.
package main

22
docker/Dockerfile Normal file
View File

@@ -0,0 +1,22 @@
FROM golang:1.23
MAINTAINER Michał Czeraszkiewicz <contact@czerasz.com>
# Set the reset cache variable
# Read more here: http://czerasz.com/2014/11/13/docker-tip-and-tricks/#use-refreshedat-variable-for-better-cache-control
ENV REFRESHED_AT 2020-09-23
# Update the package list to be able to use required packages
RUN apt-get update
# Change the working directory
WORKDIR /go/src/mgmt
# Copy all the files to the working directory
COPY . /go/src/mgmt
# Install dependencies
RUN make deps
# Build the binary
RUN make build

View File

@@ -0,0 +1,20 @@
FROM fedora:41
LABEL org.opencontainers.image.authors="laurent.indermuehle@pm.me"
ENV GOPATH=/root/gopath
ENV PATH=/root/gopath/bin:/usr/local/sbin:/sbin:/bin:/usr/sbin:/usr/bin:/usr/local/go/bin:/usr/local/bin
ENV LD_LIBRARY_PATH=/usr/lib64
ENV PKG_CONFIG_PATH=/usr/lib64/pkgconfig
# This forces make-deps.sh to install Ragel 6.1 instead of 7.0
ENV DOCKER=true
RUN dnf -y install wget unzip git make which gcc gcc-c++ ruby golang
RUN mkdir -p $GOPATH/src/github.com/purpleidea
WORKDIR $GOPATH/src/github.com/purpleidea
RUN git clone --recursive https://github.com/purpleidea/mgmt mgmt
WORKDIR $GOPATH/src/github.com/purpleidea/mgmt
RUN make deps
RUN make build
CMD ["/bin/bash"]

12
docker/Dockerfile.build Normal file
View File

@@ -0,0 +1,12 @@
FROM centos:7
MAINTAINER Karim Boumedhel <karimboumedhel@gmail.com>
ENV GOPATH=/root/gopath
ENV PATH=/opt/rh/rh-ruby22/root/usr/bin:/root/gopath/bin:/usr/local/sbin:/sbin:/bin:/usr/sbin:/usr/bin:/usr/local/go/bin
ENV LD_LIBRARY_PATH=/opt/rh/rh-ruby22/root/usr/lib64${LD_LIBRARY_PATH:+:${LD_LIBRARY_PATH}}
ENV PKG_CONFIG_PATH=/opt/rh/rh-ruby22/root/usr/lib64/pkgconfig${PKG_CONFIG_PATH:+:${PKG_CONFIG_PATH}}
RUN yum -y install epel-release wget unzip git make which centos-release-scl gcc && sed -i "s/enabled=0/enabled=1/" /etc/yum.repos.d/epel-testing.repo && yum -y install rh-ruby22 && wget -O /opt/go1.23.5.linux-amd64.tar.gz https://storage.googleapis.com/golang/go1.23.5.linux-amd64.tar.gz && tar -C /usr/local -xzf /opt/go1.23.5.linux-amd64.tar.gz
RUN mkdir -p $GOPATH/src/github.com/purpleidea && cd $GOPATH/src/github.com/purpleidea && git clone --recursive https://github.com/purpleidea/mgmt
RUN go get -u gopkg.in/alecthomas/gometalinter.v1 && cd $GOPATH/src/github.com/purpleidea/mgmt && make deps && make build
CMD ["/bin/bash"]

View File

@@ -0,0 +1,31 @@
FROM golang:1.23
MAINTAINER Michał Czeraszkiewicz <contact@czerasz.com>
# Set the reset cache variable
# Read more here: http://czerasz.com/2014/11/13/docker-tip-and-tricks/#use-refreshedat-variable-for-better-cache-control
ENV REFRESHED_AT 2019-02-06
RUN apt-get update
# Setup User to match Host User
# Give the nre user superuser permissions
ARG USER_ID=1000
ARG GROUP_ID=1000
ARG USER_NAME=mgmt
ARG GROUP_NAME=$USER_NAME
RUN groupadd --gid $GROUP_ID $GROUP_NAME && \
useradd --create-home --home /home/$USER_NAME --uid ${USER_ID} --gid $GROUP_NAME --groups sudo $USER_NAME && \
echo "$USER_NAME ALL=(ALL) NOPASSWD:ALL" >> /etc/sudoers
# Copy all the files to the working directory
COPY . /home/$USER_NAME/mgmt
# Change working directory
WORKDIR /home/$USER_NAME/mgmt
# Install dependencies
RUN make deps
# Change user
USER ${USER_NAME}

9
docker/Dockerfile.static Normal file
View File

@@ -0,0 +1,9 @@
FROM centos:7
MAINTAINER Karim Boumedhel <karimboumedhel@gmail.com>
RUN yum -y install augeas-libs libvirt-libs && yum clean all
ADD mgmt /usr/bin
RUN chmod 700 /usr/bin/mgmt
ENTRYPOINT ["/usr/bin/mgmt"]
CMD ["-h"]

26
docker/scripts/build Executable file
View File

@@ -0,0 +1,26 @@
#!/usr/bin/env bash
script_directory="$( cd "$( dirname "$0" )" && pwd )"
project_directory=$script_directory/../..
# Specify the Docker image name
image_name='purpleidea/mgmt'
# Build the image which contains the compiled binary
docker build -t $image_name \
--file=$project_directory/docker/Dockerfile $project_directory
# Remove the container if it already exists
docker rm -f mgmt-export 2> /dev/null
# Start the container in background so we can "copy out" the binary
docker run -d --name=mgmt-export $image_name bash -c 'while true; sleep 1000; done'
# Remove the current binary
rm $project_directory/mgmt 2> /dev/null
# Get the binary from the container
docker cp mgmt-export:/go/src/mgmt/mgmt $project_directory/mgmt
# Remove the container
docker rm -f mgmt-export 2> /dev/null

View File

@@ -0,0 +1,14 @@
#!/usr/bin/env bash
# Stop on any error
set -e
script_directory="$( cd "$( dirname "$0" )" && pwd )"
project_directory=$script_directory/../..
# Specify the Docker image name
image_name='purpleidea/mgmt:development'
# Build the image
docker build -t $image_name \
--file=$project_directory/docker/Dockerfile.development $project_directory

18
docker/scripts/exec-development Executable file
View File

@@ -0,0 +1,18 @@
#!/usr/bin/env bash
# runs command provided as argument inside a development (Linux) Docker container
# Stop on any error
set -e
script_directory="$( cd "$( dirname "$0" )" && pwd )"
project_directory=$script_directory/../..
# Specify the Docker image name
image_name='purpleidea/mgmt:development'
# Run container in development mode
docker run --rm --name=mgm_development --user=mgmt \
-v "$project_directory:/go/src/github.com/purpleidea/mgmt/" \
-w /go/src/github.com/purpleidea/mgmt/ \
-it "$image_name" /bin/bash -c "$*"

15
docker/scripts/run-development Executable file
View File

@@ -0,0 +1,15 @@
#!/usr/bin/env bash
# Stop on any error
set -e
script_directory="$( cd "$( dirname "$0" )" && pwd )"
project_directory=$script_directory/../..
# Specify the Docker image name
image_name='purpleidea/mgmt:development'
# Run container in development mode
docker run --rm --name=mgm_development --user=mgmt \
-v $project_directory:/home/mgmt/mgmt \
-it $image_name bash

2
docs/.gitignore vendored Normal file
View File

@@ -0,0 +1,2 @@
mgmt-documentation.pdf
_build

20
docs/Makefile Normal file
View File

@@ -0,0 +1,20 @@
# Minimal makefile for Sphinx documentation
#
# You can set these variables from the command line.
SPHINXOPTS =
SPHINXBUILD = sphinx-build
SPHINXPROJ = mgmt
SOURCEDIR = .
BUILDDIR = _build
# Put it first so that "make" without argument is like "make help".
help:
@$(SPHINXBUILD) -M help "$(SOURCEDIR)" "$(BUILDDIR)" $(SPHINXOPTS) $(O)
.PHONY: help Makefile
# Catch-all target: route all unknown targets to Sphinx using the new
# "make mode" option. $(O) is meant as a shortcut for $(SPHINXOPTS).
%: Makefile
@$(SPHINXBUILD) -M $@ "$(SOURCEDIR)" "$(BUILDDIR)" $(SPHINXOPTS) $(O)

158
docs/conf.py Normal file
View File

@@ -0,0 +1,158 @@
# -*- coding: utf-8 -*-
#
# mgmt documentation build configuration file, created by
# sphinx-quickstart on Wed Feb 15 21:34:09 2017.
#
# This file is execfile()d with the current directory set to its
# containing dir.
#
# Note that not all possible configuration values are present in this
# autogenerated file.
#
# All configuration values have a default; values that are commented out
# serve to show the default.
# If extensions (or modules to document with autodoc) are in another directory,
# add these directories to sys.path here. If the directory is relative to the
# documentation root, use os.path.abspath to make it absolute, like shown here.
#
# import os
# import sys
# sys.path.insert(0, os.path.abspath('.'))
from recommonmark.parser import CommonMarkParser
# -- General configuration ------------------------------------------------
# If your documentation needs a minimal Sphinx version, state it here.
#
# needs_sphinx = '1.0'
# Add any Sphinx extension module names here, as strings. They can be
# extensions coming with Sphinx (named 'sphinx.ext.*') or your custom
# ones.
extensions = []
# Add any paths that contain templates here, relative to this directory.
templates_path = ['_templates']
# The suffix(es) of source filenames.
# You can specify multiple suffix as a list of string:
#
source_parsers = {
'.md': CommonMarkParser,
}
source_suffix = ['.rst', '.md']
# The master toctree document.
master_doc = 'index'
# General information about the project.
project = u'mgmt'
copyright = u'Copyright (C) James Shubin and the project contributors'
author = u'James Shubin'
# The version info for the project you're documenting, acts as replacement for
# |version| and |release|, also used in various other places throughout the
# built documents.
#
# The short X.Y version.
version = u''
# The full version, including alpha/beta/rc tags.
release = u''
# The language for content autogenerated by Sphinx. Refer to documentation
# for a list of supported languages.
#
# This is also used if you do content translation via gettext catalogs.
# Usually you set "language" from the command line for these cases.
language = None
# List of patterns, relative to source directory, that match files and
# directories to ignore when looking for source files.
# This patterns also effect to html_static_path and html_extra_path
exclude_patterns = ['_build', 'Thumbs.db', '.DS_Store', 'venv']
# The name of the Pygments (syntax highlighting) style to use.
pygments_style = 'sphinx'
# If true, `todo` and `todoList` produce output, else they produce nothing.
todo_include_todos = False
# -- Options for HTML output ----------------------------------------------
# The theme to use for HTML and HTML Help pages. See the documentation for
# a list of builtin themes.
#
#html_theme = 'alabaster'
# Theme options are theme-specific and customize the look and feel of a theme
# further. For a list of options available for each theme, see the
# documentation.
#
# html_theme_options = {}
# Add any paths that contain custom static files (such as style sheets) here,
# relative to this directory. They are copied after the builtin static files,
# so a file named "default.css" will overwrite the builtin "default.css".
html_static_path = ['_static']
# -- Options for HTMLHelp output ------------------------------------------
# Output file base name for HTML help builder.
htmlhelp_basename = 'mgmtdoc'
# -- Options for LaTeX output ---------------------------------------------
latex_elements = {
# The paper size ('letterpaper' or 'a4paper').
#
# 'papersize': 'letterpaper',
# The font size ('10pt', '11pt' or '12pt').
#
# 'pointsize': '10pt',
# Additional stuff for the LaTeX preamble.
#
# 'preamble': '',
# Latex figure (float) alignment
#
# 'figure_align': 'htbp',
}
# Grouping the document tree into LaTeX files. List of tuples
# (source start file, target name, title,
# author, documentclass [howto, manual, or own class]).
latex_documents = [
(master_doc, 'mgmt.tex', u'mgmt Documentation',
u'James Shubin', 'manual'),
]
# -- Options for manual page output ---------------------------------------
# One entry per manual page. List of tuples
# (source start file, name, description, authors, manual section).
man_pages = [
(master_doc, 'mgmt', u'mgmt Documentation',
[author], 1)
]
# -- Options for Texinfo output -------------------------------------------
# Grouping the document tree into Texinfo files. List of tuples
# (source start file, target name, title, author,
# dir menu entry, description, category)
texinfo_documents = [
(master_doc, 'mgmt', u'mgmt Documentation',
author, 'mgmt', 'Next generation distributed, event-driven, parallel config management!',
'Miscellaneous'),
]

96
docs/contributing.md Normal file
View File

@@ -0,0 +1,96 @@
# Contributing
What follows is a short guide with information for participants who wish to
contribute to the project. It hopes to set both some expectations and boundaries
so that we both benefit.
## Small patches
If you have a small patch which you believe is straightforward, should be easy
to merge, and isn't overly onerous on your time to write, please feel free to
send it our way without asking first. Bug fixes are excellent examples of small
patches. Please make sure to familiarize yourself with the rough coding style of
the project first, and read through the [style guide](style-guide.md).
## Making an excellent small patch
As a special case: We'd like to avoid minimal effort, one-off, drive-by patches
by bots and contributors looking to increase their "activity" numbers. As an
example: a patch which fixes a small linting issue isn't rousing, but a patch
that adds a linter test _and_ fixes a small linting issue is, because it shows
you put in more effort.
## Medium patches
Medium sized patches are especially welcome. Good examples of these patches
can include writing a new `mgmt` resource or function. You'll generally need
some knowledge of golang interfaces and concurrency to write these patches.
Before writing one of these, please make sure you understand some basics about
the project and how the tool works. After this, it is recommended that you join
our discussion channel to suggest the idea, and ideally include the actual API
you'd like to propose before writing the code and sending a patch.
## Making an excellent medium patch proposal
The "API" of a resource is the type signature of the resource struct, and the
"API" of a function is the type signature or signatures that it supports. (Since
functions can be polymorphic, more than one signature can be possible!) A good
proposal would likely also comment on the mechanisms the resources or functions
would use to watch for events, to check state, and to apply changes. If these
mechanisms need new dependencies, a brief survey of which dependencies are
available and why you recommend a particular one is encouraged.
## Large patches or structural and core patches
Please do not send us large, core or structurally significant patches without
first getting our approval and without getting some medium patches in first.
These patches take a lot of effort to review, and we don't want to skimp on our
commitment to that if we can't muster it. Instead grow our relationship with you
on the medium-sized patches first. (A core patch might refer to something that
touches either the function engine, resource engine, compiler internals, or
something that is part of one of the internal API's.)
## Expectations and boundaries
When interacting with the project and soliciting feedback (either for design or
during a code review) please keep in mind that the project (unfortunately!) has
time constraints and so must prioritize how it handles workloads. If you are
someone who has successfully sent in small patches, we will be more willing to
spend time mentoring your medium sized patches and so on. Think of it this way:
as you show that you're contributing to the project, we'll contribute more to
you. Put another way: we can't afford to spend large amounts of time discussing
potential patches with you, just to end up nowhere. Build up your reputation
with us, and we hope to help grow our symbiosis with you all the while as you
grow too!
## Energy output
The same goes for users and issue creators. There are times when we simply don't
have the cycles to discuss or litigate an issue with you. We wish we did have
more time, but it is finite, and running a project is not free. Therefore,
please keep in mind that you don't automatically qualify for free support or
attention.
## Attention seeking behaviours
Some folks spend too much time starting discussions, commenting on issues,
"planning" and otherwise displaying attention seeking behaviours. Please avoid
doing this as much as possible, especially if you are not already a major
contributor to the project. While it may be well intentioned, if it is
indistinguishable to us from intentional interference, then it's not welcome
behaviour. Remember that Free Software is not free to write. If you require more
attention, then either contribute more to the project, or consider paying for a
[support contract](https://mgmtconfig.com/).
## Consulting
Having said all that, there are some folks who want to do some longer-term
planning to decide if our core design and architecture is right for them to
invest in. If that's the case, and you aren't already a well-known project
contributor, please [contact](https://mgmtconfig.com/) us for a consulting
quote. We have packages available for both individuals and businesses.
## Respect
Please be mindful and respectful of others when interacting with the project and
its contributors. If you cannot abide by that, you may no longer be welcome.

163
docs/development.md Normal file
View File

@@ -0,0 +1,163 @@
# Development
This document contains some additional information and help regarding
developing `mgmt`. Useful tools, conventions, etc.
Be sure to read [quick start guide](quick-start-guide.md) first.
## Vagrant
If you would like to avoid doing the above steps manually, we have prepared a
[Vagrant](https://www.vagrantup.com/) environment for your convenience. From the
project directory, run a `vagrant up`, and then a `vagrant status`. From there,
you can `vagrant ssh` into the `mgmt` machine. The `MOTD` will explain the rest.
This environment isn't commonly used by the `mgmt` developers, so it might not
be working properly.
## Using Docker
Alternatively, you can check out the [docker folder](../docker/) in order to
develop or deploy using docker. This method is not endorsed or supported, so use
at your own risk, as it might not be working properly.
## Information about dependencies
Software projects have a few different kinds of dependencies. There are _build_
dependencies, _runtime_ dependencies, and additionally, a few extra dependencies
required for running the _test_ suite.
### Build
* A modern `golang` version. The version available in the current Fedora
releases is usually supported. This is also distributed as a binary officially
by [golang.org](https://golang.org/dl/).
### Runtime
A relatively modern GNU/Linux system should be able to run `mgmt` without any
problems. Since `mgmt` runs as a single statically compiled binary, all of the
library dependencies are included. It is expected, that certain advanced
resources require host specific facilities to work. These requirements are
listed below:
| Resource | Dependency | Version | Check version with |
|----------|-------------------|-----------------------------|-----------------------------------------------------------|
| augeas | augeas-devel | `augeas 1.6` or greater | `dnf info augeas-devel` or `apt-cache show libaugeas-dev` |
| file | inotify | `Linux 2.6.27` or greater | `uname -a` |
| hostname | systemd-hostnamed | `systemd 25` or greater | `systemctl --version` |
| nspawn | systemd-nspawn | `systemd ???` or greater | `systemctl --version` |
| pkg | packagekitd | `packagekit 1.x` or greater | `pkcon --version` |
| svc | systemd | `systemd ???` or greater | `systemctl --version` |
| virt | libvirt-devel | `libvirt 1.2.0` or greater | `dnf info libvirt-devel` or `apt-cache show libvirt-dev` |
| virt | libvirtd | `libvirt 1.2.0` or greater | `libvirtd --version` |
For building a visual representation of the graph, `graphviz` is required.
To build `mgmt` without augeas support please run:
`GOTAGS='noaugeas' make build`
To build `mgmt` without libvirt support please run:
`GOTAGS='novirt' make build`
To build `mgmt` without docker support please run:
`GOTAGS='nodocker' make build`
To build `mgmt` without augeas, libvirt or docker support please run:
`GOTAGS='noaugeas novirt nodocker' make build`
## OSX/macOS/Darwin development
Developing and running `mgmt` on macOS is currently not supported (but not
discouraged either). Meaning it might work but in the case it doesn't you would
have to provide your own patches to fix problems (the project maintainer and
community are glad to assist where needed).
There are currently some issues that make `mgmt` less suitable to run for
provisioning macOS. But as a client to provision remote servers it should run
fine.
Since the primary supported systems are Linux and these are the environments
tested, it is wise to run these suites during macOS development as well. To ease
this, Docker can be leveraged ([Docker for Mac](https://docs.docker.com/docker-for-mac/)).
Before running any of the commands below create the development Docker image:
```
docker/scripts/build-development
```
This image requires updating every time dependencies (`make-deps.sh`) changes.
Then to run the test suite:
```
docker run --rm -ti \
-v $PWD:/go/src/github.com/purpleidea/mgmt/ \
-w /go/src/github.com/purpleidea/mgmt/ \
purpleidea/mgmt:development \
make test
```
For convenience this command is wrapped in `docker/scripts/exec-development`.
Basically any command can be executed this way. Because the repository source is
mounted into the Docker container invocation will be quick and allow rapid
testing, for example:
```
docker/scripts/exec-development test/test-shell.sh load0.sh
```
Other examples:
```
docker/scripts/exec-development make build
docker/scripts/exec-development ./mgmt run --tmp-prefix lang examples/lang/load0.mcl
```
Be advised that this method is not supported and it might not be working
properly.
## Testing
This project has both unit tests in the form of golang tests and integration
tests using shell scripting.
Native golang tests are preferred over tests written in our shell testing
framework. Please see [https://golang.org/pkg/testing/](https://golang.org/pkg/testing/)
for more information.
To run all tests:
```
make test
```
There is a library of quick and small integration tests for the language and
YAML related things, check out [`test/shell/`](/test/shell). Adding a test is as
easy as copying one of the files in [`test/shell/`](/test/shell) and adapting
it.
This test suite won't run by default (unless when on CI server) and needs to be
called explicitly using:
```
make test-shell
```
Or run an individual shell test using:
```
make test-shell-load0
```
Tip: you can use TAB completion with `make` to quickly get a list of possible
individual tests to run.
## Tools, integrations, IDE's etc
### IDE/Editor support
* Emacs: see `misc/emacs/`
* [Textmate](https://github.com/aequitas/mgmt.tmbundle)
* [VSCode](https://github.com/aequitas/mgmt.vscode)

50
docs/docs.go Normal file
View File

@@ -0,0 +1,50 @@
// Mgmt
// Copyright (C) James Shubin and the project contributors
// Written by James Shubin <james@shubin.ca> and the project contributors
//
// This program is free software: you can redistribute it and/or modify
// it under the terms of the GNU General Public License as published by
// the Free Software Foundation, either version 3 of the License, or
// (at your option) any later version.
//
// This program is distributed in the hope that it will be useful,
// but WITHOUT ANY WARRANTY; without even the implied warranty of
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
// GNU General Public License for more details.
//
// You should have received a copy of the GNU General Public License
// along with this program. If not, see <https://www.gnu.org/licenses/>.
//
// Additional permission under GNU GPL version 3 section 7
//
// If you modify this program, or any covered work, by linking or combining it
// with embedded mcl code and modules (and that the embedded mcl code and
// modules which link with this program, contain a copy of their source code in
// the authoritative form) containing parts covered by the terms of any other
// license, the licensors of this program grant you additional permission to
// convey the resulting work. Furthermore, the licensors of this program grant
// the original author, James Shubin, additional permission to update this
// additional permission if he deems it necessary to achieve the goals of this
// additional permission.
// Package docs provides a tool that generates documentation from the source.
//
// ./mgmt docs generate --output /tmp/docs.json && cat /tmp/docs.json | jq
package docs
import (
"context"
)
// API is the simple interface we expect for any setup items.
type API interface {
// Main runs everything for this setup item.
Main(context.Context) error
}
// Config is a struct of all the configuration values which are shared by all of
// the setup utilities. By including this as a separate struct, it can be used
// as part of the API if we want.
type Config struct {
//Foo string `arg:"--foo,env:MGMT_DOCGEN_FOO" help:"Foo..."` // TODO: foo
}

547
docs/documentation.md Normal file
View File

@@ -0,0 +1,547 @@
# General documentation
## Overview
The `mgmt` tool is a next generation config management solution. It's ready for
production, and we hope you try out the future soon. Get involved today!
## Project Description
The mgmt tool is a distributed, event driven, config management tool, that
supports parallel execution, and librarification to be used as the management
foundation in and for, new and existing software.
For more information, you may like to read some blog posts from the author:
* [Next generation config mgmt](https://purpleidea.com/blog/2016/01/18/next-generation-configuration-mgmt/)
* [Automatic edges in mgmt](https://purpleidea.com/blog/2016/03/14/automatic-edges-in-mgmt/)
* [Automatic grouping in mgmt](https://purpleidea.com/blog/2016/03/30/automatic-grouping-in-mgmt/)
* [Automatic clustering in mgmt](https://purpleidea.com/blog/2016/06/20/automatic-clustering-in-mgmt/)
* [Remote execution in mgmt](https://purpleidea.com/blog/2016/10/07/remote-execution-in-mgmt/)
* [Send/Recv in mgmt](https://purpleidea.com/blog/2016/12/07/sendrecv-in-mgmt/)
* [Metaparameters in mgmt](https://purpleidea.com/blog/2017/03/01/metaparameters-in-mgmt/)
There is also an [introductory video](https://www.youtube.com/watch?v=LkEtBVLfygE&html5=1)
available. Older videos and other material [is available](on-the-web.md).
## Setup
You'll probably want to read the [quick start guide](quick-start-guide.md) to
get going.
## Features
This section details the numerous features of mgmt and some caveats you might
need to be aware of.
### Autoedges
Automatic edges, or AutoEdges, is the mechanism in mgmt by which it will
automatically create dependencies for you between resources. For example,
since mgmt can discover which files are installed by a package it will
automatically ensure that any file resource you declare that matches a
file installed by your package resource will only be processed after the
package is installed.
#### Controlling autoedges
Though autoedges is likely to be very helpful and avoid you having to declare
all dependencies explicitly, there are cases where this behaviour is
undesirable.
Some distributions allow package installations to automatically start the
service they ship. This can be problematic in the case of packages like MySQL
as there are configuration options that need to be set before MySQL is ever
started for the first time (or you'll need to wipe the data directory). In
order to handle this situation you can disable autoedges per resource and
explicitly declare that you want `my.cnf` to be written to disk before the
installation of the `mysql-server` package.
You can disable autoedges for a resource by setting the `autoedge` key on
the meta attributes of that resource to `false`.
#### Blog post
You can read the introductory blog post about this topic here:
[https://purpleidea.com/blog/2016/03/14/automatic-edges-in-mgmt/](https://purpleidea.com/blog/2016/03/14/automatic-edges-in-mgmt/)
### Autogrouping
Automatic grouping or AutoGroup is the mechanism in mgmt by which it will
automatically group multiple resource vertices into a single one. This is
particularly useful for grouping multiple package resources into a single
resource, since the multiple installations can happen together in a single
transaction, which saves a lot of time because package resources typically have
a large fixed cost to running (downloading and verifying the package repo) and
if they are grouped they share this fixed cost. This grouping feature can be
used for other use cases too.
You can disable autogrouping for a resource by setting the `autogroup` key on
the meta attributes of that resource to `false`.
#### Blog post
You can read the introductory blog post about this topic here:
[https://purpleidea.com/blog/2016/03/30/automatic-grouping-in-mgmt/](https://purpleidea.com/blog/2016/03/30/automatic-grouping-in-mgmt/)
### Automatic clustering
Automatic clustering is a feature by which mgmt automatically builds, scales,
and manages the embedded etcd cluster which is compiled into mgmt itself. It is
quite helpful for rapidly bootstrapping clusters and avoiding the extra work to
setup etcd.
If you prefer to avoid this feature. you can always opt to use an existing etcd
cluster that is managed separately from mgmt by pointing your mgmt agents at it
with the `--seeds` variable.
#### Blog post
You can read the introductory blog post about this topic here:
[https://purpleidea.com/blog/2016/06/20/automatic-clustering-in-mgmt/](https://purpleidea.com/blog/2016/06/20/automatic-clustering-in-mgmt/)
### Remote ("agent-less") mode
Remote mode is a special mode that lets you kick off mgmt runs on one or more
remote machines which are only accessible via SSH. In this mode the initiating
host connects over SSH, copies over the `mgmt` binary, opens an SSH tunnel, and
runs the remote program while simultaneously passing the etcd traffic back
through the tunnel so that the initiators etcd cluster can be used to exchange
resource data.
The interesting benefit of this architecture is that multiple hosts which can't
connect directly use the initiator to pass the important traffic through to each
other. Once the cluster has converged all the remote programs can shutdown
leaving no residual agent.
This mode can also be useful for bootstrapping a new host where you'd like to
have the service run continuously and as part of an mgmt cluster normally.
In particular, when combined with the `--converged-timeout` parameter, the
entire set of running mgmt agents will need to all simultaneously converge for
the group to exit. This is particularly useful for bootstrapping new clusters
which need to exchange information that is only available at run time.
This existed in earlier versions of mgmt as a `--remote` option, but it has been
removed and is being ported to a more powerful variant where you can remote
execute via a `remote` resource.
#### Blog post
You can read the introductory blog post about this topic here:
[https://purpleidea.com/blog/2016/10/07/remote-execution-in-mgmt/](https://purpleidea.com/blog/2016/10/07/remote-execution-in-mgmt/)
### Puppet support
You can supply a puppet manifest instead of creating the (YAML) graph manually.
Puppet must be installed and in `mgmt`'s search path. You also need the
[ffrank-mgmtgraph puppet module](https://forge.puppet.com/ffrank/mgmtgraph).
Invoke `mgmt` with the `--puppet` switch, which supports 3 variants:
1. Request the configuration from the puppet server (like `puppet agent` does)
`mgmt run puppet --puppet agent`
2. Compile a local manifest file (like `puppet apply`)
`mgmt run puppet --puppet /path/to/my/manifest.pp`
3. Compile an ad hoc manifest from the commandline (like `puppet apply -e`)
`mgmt run puppet --puppet 'file { "/etc/ntp.conf": ensure => file }'`
For more details and caveats see [puppet-guide.md](puppet-guide.md).
#### Blog post
An introductory post on the puppet support is on
[Felix's blog](http://ffrank.github.io/features/2016/06/19/puppet-powered-mgmt/).
## Reference
Please note that there are a number of undocumented options. For more
information on these options, please view the source at:
[https://github.com/purpleidea/mgmt/](https://github.com/purpleidea/mgmt/).
If you feel that a well used option needs documenting here, please patch it!
### Overview of reference
* [Meta parameters](#meta-parameters): List of available resource meta parameters.
* [Lang metadata file](#lang-metadata-file): Lang metadata file format.
* [Graph definition file](#graph-definition-file): Main graph definition file.
* [Command line](#command-line): Command line parameters.
* [Compilation options](#compilation-options): Compilation options.
### Meta parameters
These meta parameters are special parameters (or properties) which can apply to
any resource. The usefulness of doing so will depend on the particular meta
parameter and resource combination.
#### AutoEdge
Boolean. Should we generate auto edges for this resource?
#### AutoGroup
Boolean. Should we attempt to automatically group this resource with others?
#### Noop
Boolean. Should the Apply portion of the CheckApply method of the resource
make any changes? Noop is a concatenation of no-operation.
#### Retry
Integer. The number of times to retry running the resource on error. Use -1 for
infinite. This currently applies for both the Watch operation (which can fail)
and for the CheckApply operation. While they could have separate values, I've
decided to use the same ones for both until there's a proper reason to want to
do something differently for the Watch errors.
#### Delay
Integer. Number of milliseconds to wait between retries. The same value is
shared between the Watch and CheckApply retries. This currently applies for both
the Watch operation (which can fail) and for the CheckApply operation. While
they could have separate values, I've decided to use the same ones for both
until there's a proper reason to want to do something differently for the Watch
errors.
#### Poll
Integer. Number of seconds to wait between `CheckApply` checks. If this is
greater than zero, then the standard event based `Watch` mechanism for this
resource is replaced with a simple polling mechanism. In general, this is not
recommended, unless you have a very good reason for doing so.
Please keep in mind that if you have a resource which changes every `I` seconds,
and you poll it every `J` seconds, and you've asked for a converged timeout of
`K` seconds, and `I <= J <= K`, then your graph will likely never converge.
When polling, the system detects that a resource is not converged if its
`CheckApply` method returns false. This allows a resource which changes every
`I` seconds, and which is polled every `J` seconds, and with a converged timeout
of `K` seconds to still converge when `J <= K`, as long as `I > J || I > K`,
which is another way of saying that if the resource finally settles down to give
the graph enough time, it can probably converge.
#### Limit
Float. Maximum rate of `CheckApply` runs started per second. Useful to limit
an especially _eventful_ process from causing excessive checks to run. This
defaults to `+Infinity` which adds no limiting. If you change this value, you
will also need to change the `Burst` value to a non-zero value. Please see the
[rate](https://godoc.org/golang.org/x/time/rate) package for more information.
#### Burst
Integer. Burst is the maximum number of runs which can happen without invoking
the rate limiter as designated by the `Limit` value. If the `Limit` is not set
to `+Infinity`, this must be a non-zero value. Please see the
[rate](https://godoc.org/golang.org/x/time/rate) package for more information.
#### Sema
List of string ids. Sema is a P/V style counting semaphore which can be used to
limit parallelism during the CheckApply phase of resource execution. Each
resource can have `N` different semaphores which share a graph global namespace.
Each semaphore has a maximum count associated with it. The default value of the
size is 1 (one) if size is unspecified. Each string id is the unique id of the
semaphore. If the id contains a trailing colon (:) followed by a positive
integer, then that value is the max size for that semaphore. Valid semaphore
id's include: `some_id`, `hello:42`, `not:smart:4` and `:13`. It is expected
that the last bare example be only used by the engine to add a global semaphore.
#### Rewatch
Boolean. Rewatch specifies whether we re-run the Watch worker during a graph
swap if it has errored. When doing a graph compare to swap the graphs, if this
is true, and this particular worker has errored, then we'll remove it and add it
back as a new vertex, thus causing it to run again. This is different from the
`Retry` metaparam which applies during the normal execution. It is only when
this is exhausted that we're in permanent worker failure, and only then can we
rely on this metaparam.
#### Realize
Boolean. Realize ensures that the resource is guaranteed to converge at least
once before a potential graph swap removes or changes it. This guarantee is
useful for fast changing graphs, to ensure that the brief creation of a resource
is seen. This guarantee does not prevent against the engine quitting normally,
and it can't guarantee it if the resource is blocked because of a failed
pre-requisite resource.
*XXX: This is currently not implemented!*
#### Dollar
Boolean. Dollar allows you to have a resource name that starts with a `$` sign.
This is false by default. This helps you catch cases when you write code like:
```mcl
$foo = "/tmp/file1"
file "$foo" {} # incorrect!
```
The above code would ignore the `$foo` variable and attempt to make a file named
`$foo` which would obviously not work. To correctly interpolate a variable, you
need to surround the name with curly braces.
```mcl
$foo = "/tmp/file1"
file "${foo}" {} # correct!
```
This meta param is a safety measure to make your life easier. It works for all
resources. If someone comes up with a resource which would routinely start with
a dollar sign, then we can revisit the default for this resource kind.
#### Hidden
Boolean. Hidden means that this resource will not get executed on the resource
graph on which it is defined. This can be used as a simple boolean switch, or,
more commonly in combination with the Export meta param which specifies that the
resource params are exported into the shared database. When this is true, it
does not prevent export. In fact, it is commonly used in combination with
Export. Using this option will still include it in the resource graph, but it
will exist there in a special "mode" where it will not conflict with any other
identically named resources. It can even be used as part of an edge or via a
send/recv receiver. It can NOT be a sending vertex. These properties
differentiate the use of this instead of simply wrapping a resource in an "if"
statement.
#### Export
List of strings. Export is a list of hostnames (and/or the special "*" entry)
which if set, will mark this resource data as intended for export to those
hosts. This does not prevent any users of the shared data storage from reading
these values, so if you want to guarantee secrecy, use the encryption
primitives. This only labels the data accordingly, so that other hosts can know
what data is available for them to collect. The (kind, name, host) export triple
must be unique from any given exporter. In other words, you may not export two
different instances of a kind+name to the same host, the exports must not
conflict. On resource collect, this parameter is not preserved.
```mcl
file "/tmp/foo" {
state => "exists",
content => "i'm exported!\n",
Meta:hidden => true,
Meta:export => ["h1",],
}
file "/tmp/foo" {
state => "exists",
content => "i'm exported AND i'm used here\n",
Meta:export => ["h1",],
}
```
#### Reverse
Boolean. Reverse is a property that some resources can implement that specifies
that some "reverse" operation should happen when that resource "disappears". A
disappearance happens when a resource is defined in one instance of the graph,
and is gone in the subsequent one. This disappearance can happen if it was
previously in an if statement that then becomes false.
This is helpful for building robust programs with the engine. The engine adds a
"reversed" resource to that subsequent graph to accomplish the desired "reverse"
mechanics. The specifics of what this entails is a property of the particular
resource that is being "reversed".
It might be wise to combine the use of this meta parameter with the use of the
`realize` meta parameter to ensure that your reversed resource actually runs at
least once, if there's a chance that it might be gone for a while.
### Lang metadata file
Any module *must* have a metadata file in its root. It must be named
`metadata.yaml`, even if it's empty. You can specify zero or more values in yaml
format which can change how your module behaves, and where the `mcl` language
looks for code and other files. The most important top level keys are: `main`,
`path`, `files`, and `license`.
#### Main
The `main` key points to the default entry point of your code. It must be a
relative path if specified. If it's empty it defaults to `main.mcl`. It should
generally not be changed. It is sometimes set to `main/main.mcl` if you'd like
your modules code out of the root and into a child directory for cases where you
don't plan on having a lot deeper imports relative to `main.mcl` and all those
files would clutter things up.
#### Path
The `path` key specifies the modules import search directory to use for this
module. You can specify this if you'd like to vendor something for your module.
In general, if you use it, please use the convention: `path/`. If it's not
specified, you will default to the parent modules directory.
#### Files
The `files` key specifies some additional files that will get included in your
deploy. It defaults to `files/`.
#### License
The `license` key allows you to specify a license for the module. Please specify
one so that everyone can enjoy your code! Use a "short license identifier", like
`LGPLv3+`, or `MIT`. The former is a safe choice if you're not sure what to use.
### Graph definition file
graph.yaml is the compiled graph definition file. The format is currently
undocumented, but by looking through the [examples/](https://github.com/purpleidea/mgmt/tree/master/examples/yaml/)
you can probably figure out most of it, as it's fairly intuitive. It's not
recommended that you use this, since it's preferable to write code in the
[mcl language](language-guide.md) front-end.
### Command line
The main interface to the `mgmt` tool is the command line. For the most recent
documentation, please run `mgmt --help`.
#### `--converged-timeout <seconds>`
Exit if the machine has converged for approximately this many seconds.
#### `--max-runtime <seconds>`
Exit when the agent has run for approximately this many seconds. This is not
generally recommended, but may be useful for users who know what they're doing.
#### `--noop`
Globally force all resources into no-op mode. This also disables the export to
etcd functionality, but does not disable resource collection, however all
resources that are collected will have their individual noop settings set.
#### `--sema <size>`
Globally add a counting semaphore of this size to each resource in the graph.
The semaphore will get given an id of `:size`. In other words if you specify a
size of 42, you can expect a semaphore if named: `:42`. It is expected that
consumers of the semaphore metaparameter always include a prefix to avoid a
collision with this globally defined semaphore. The size value must be greater
than zero at this time. The traditional non-parallel execution found in config
management tools such as `puppet` can be obtained with `--sema 1`.
#### `--ssh-priv-id-rsa`
Specify the path for finding SSH keys. This defaults to `~/.ssh/id_rsa`. To
never use this method of authentication, set this to the empty string.
#### `--prefix <path>`
Specify a path to a custom working directory prefix. This directory will get
created if it does not exist. This usually defaults to `/var/lib/mgmt/`. This
can't be combined with the `--tmp-prefix` option. It can be combined with the
`--allow-tmp-prefix` option.
#### `--tmp-prefix`
If this option is specified, a temporary prefix will be used instead of the
default prefix. This can't be combined with the `--prefix` option.
#### `--allow-tmp-prefix`
If this option is specified, we will attempt to fall back to a temporary prefix
if the primary prefix couldn't be created. This is useful for avoiding failures
in environments where the primary prefix may or may not be available, but you'd
like to try. The canonical example is when running `mgmt` with remote execution
there might be a cached copy of the binary in the primary prefix, but if there's
no binary available continue working in a temporary directory to avoid failure.
### Compilation options
You can control some compilation variables by using environment variables.
#### Disable libvirt support
If you wish to compile mgmt without libvirt, you can use the following command:
```
GOTAGS=novirt make build
```
#### Disable augeas support
If you wish to compile mgmt without augeas support, you can use the following
command:
```
GOTAGS=noaugeas make build
```
#### Disable docker support
If you wish to compile mgmt without docker support, you can use the following
command:
```
GOTAGS=nodocker make build
```
#### Combining compile-time flags
You can combine multiple tags by using a space-separated list:
```
GOTAGS="noaugeas novirt nodocker" make build
```
## Examples
For example configurations, please consult the [examples/](https://github.com/purpleidea/mgmt/tree/master/examples)
directory in the git source repository. It is available from:
[https://github.com/purpleidea/mgmt/tree/master/examples](https://github.com/purpleidea/mgmt/tree/master/examples)
### Systemd:
See [`misc/mgmt.service`](../misc/mgmt.service) for a sample systemd unit file.
This unit file is part of the RPM.
To specify your custom options for `mgmt` on a systemd distro:
```bash
sudo mkdir -p /etc/systemd/system/mgmt.service.d/
cat > /etc/systemd/system/mgmt.service.d/env.conf <<EOF
# Environment variables:
MGMT_SEEDS=http://127.0.0.1:2379
MGMT_CONVERGED_TIMEOUT=-1
MGMT_MAX_RUNTIME=0
# Other CLI options if necessary.
#OPTS="--max-runtime=0"
EOF
sudo systemctl daemon-reload
```
## Development
This is a project that I started in my free time in 2013. Development is driven
by all of our collective patches! Dive right in, and start hacking!
Please contact me if you'd like to invite me to speak about this at your event.
You can follow along [on my technical blog](https://purpleidea.com/blog/).
To report any bugs, please file a ticket at: [https://github.com/purpleidea/mgmt/issues](https://github.com/purpleidea/mgmt/issues).
## Authors
Copyright (C) James Shubin and the project contributors
Please see the
[AUTHORS](https://github.com/purpleidea/mgmt/tree/master/AUTHORS) file
for more information.
* [github](https://github.com/purpleidea/)
* [&#64;purpleidea](https://twitter.com/#!/purpleidea)
* [https://purpleidea.com/](https://purpleidea.com/)

537
docs/faq.md Normal file
View File

@@ -0,0 +1,537 @@
## Frequently asked questions
(Send your questions as a patch to this FAQ! I'll review it, merge it, and
respond by commit with the answer.)
### Why did you start this project?
I wanted a next generation config management solution that didn't have all of
the design flaws or limitations that the current generation of tools do, and no
tool existed!
### Why did you choose `golang` for the project?
When I started working on the project, I needed to choose a language that
already had an implementation of a distributed consensus algorithm available.
That meant [Paxos](https://en.wikipedia.org/wiki/Paxos_(computer_science)) or
[Raft](https://en.wikipedia.org/wiki/Raft_(computer_science)). Golang was one
language that actually had two different Raft implementations, `etcd`, and
`consul`. Other design requirements included something that was reasonably fast,
typed and memory-safe, and suited for systems engineering. After a reasonably
extensive search, I chose `golang`. I think it was the right decision. There are
a number of other features of the language which helped influence the decision.
### How do I contribute to the project if I don't know `golang`?
There are many different ways you can contribute to the project. They can be
broadly divided into two main categories:
1. With contributions written in `golang`
2. With contributions _not_ written in `golang`
If you do not know `golang`, and have no desire to learn, you can still
contribute to mgmt by using it, testing it, writing docs, or even just by
telling your friends about it. If you don't mind some coding, learning about the
[mgmt language](https://purpleidea.com/blog/2018/02/05/mgmt-configuration-language/)
might be an enjoyable experience for you. It is a small [DSL](https://en.wikipedia.org/wiki/Domain-specific_language)
and not a general purpose programming language, and you might find it more fun
than what you're typically used to. One of the reasons the mgmt author got into
writing automation modules, was because he found it much more fun to build with
a higher level DSL, than in a general purpose programming language.
If you do not know `golang`, and would like to learn, are a beginner and want to
improve your skills, or want to gain some great interdisciplinary systems
engineering knowledge around a cool automation project, we're happy to mentor
you. Here are some pre-requisites steps which we recommend:
1. Make sure you have a somewhat recent GNU/Linux environment to hack on. A
recent [Fedora](https://getfedora.org/) or [Debian](https://www.debian.org/)
environment is recommended. Developing, testing, and contributing on `macOS` or
`Windows` will be either more difficult or impossible.
2. Ensure that you're mildly comfortable with the basics of using `git`. You can
find a number of tutorials online.
3. Spend between four to six hours with the [golang tour](https://tour.golang.org/).
Skip over the longer problems, but try and get a solid overview of everything.
If you forget something, you can always go back and repeat those parts.
4. Connect to our [#mgmtconfig](https://matrix.to/#/#mgmtconfig:matrix.org)
Matrix channel and hang out with us there out there.
5. Now it's time to try and starting writing a patch! We have tagged a bunch of
[open issues as #mgmtlove](https://github.com/purpleidea/mgmt/issues?q=is%3Aissue+is%3Aopen+label%3Amgmtlove)
for new users to have somewhere to get involved. Look through them to see if
something interests you. If you find one, let us know you're working on it by
leaving a comment in the ticket. We'll be around to answer questions in the
channel, and to create new issues if there wasn't something that fit your
interests. When you submit a patch, we'll review it and give you some feedback.
Over time, we hope you'll learn a lot while supporting the project! Now get
hacking!
### Is this project ready for production?
It's getting pretty close. I'm able to write modules for it now!
Compared to some existing automation tools out there, mgmt is a relatively new
project. It is probably not as feature complete as some other software, but it
also offers a number of features which are not currently available elsewhere.
Because we have not released a `1.0` release yet, we are not guaranteeing
stability of the internal or external API's. We only change them if it's really
necessary, and we don't expect anything particularly drastic to occur. We would
expect it to be relatively easy to adapt your code if such changes happened.
As with all software, bugs can occur, and while we make no guarantees of being
bug-free, there are a number of things we've done to reduce the chances of one
causing you trouble:
1. Our software is written in golang, which is a memory-safe language, and which
is known to reduce or eliminate entire classes of bugs.
2. We have a test suite which we run on every commit, and every 24 hours. If you
have a particular case that you'd like to test, you are welcome to add it in!
3. The mgmt language itself offers a number of safety features. You can
[read about them in the introductory blog post](https://purpleidea.com/blog/2018/02/05/mgmt-configuration-language/).
Having said all this, as with all software, there are still missing features
which some users might want in their production environments. We're working hard
to get all of those implemented, but we hope that you'll get involved and help
us finish off the ones that are most important to you. We are happy to mentor
new contributors, and have even [tagged](https://github.com/purpleidea/mgmt/issues?q=is%3Aissue+is%3Aopen+label%3Amgmtlove)
a number of issues if you need help getting started.
Some of the current limitations include:
* Auth hasn't been implemented yet, so you should only use it in trusted
environments (not on publicly accessible networks) for now.
* The number of built-in core functions is still small. You may encounter
scenarios where you're missing a function. The good news is that it's relatively
easy to add this missing functionality yourself. In time, with your help, the
list will grow!
* Large file distribution is not yet implemented. You might want a scenario
where mgmt is used to distribute large files (such as `.iso` images) throughout
your cluster. While this isn't a common use-case, it won't be possible until
someone wants to write the patch. (Mentoring available!) You can workaround this
easily by storing those files on a separate fileserver for the interim.
* There isn't an ecosystem of community `modules` yet. We've got this on our
roadmap, so please stay tuned!
We hope you'll participate as an early adopter. Every additional pair of helping
hands gets us all there faster! It's quite possible to use this to build useful
automation today, and we hope you'll start getting familiar with the software.
### Why did you use etcd? What about consul?
Etcd and consul are both written in golang, which made them the top two
contenders for my prototype. Ultimately a choice had to be made, and etcd was
chosen, but it was also somewhat arbitrary. If there is available interest,
good reasoning, *and* patches, then we would consider either switching or
supporting both, but this is not a high priority at this time.
### Can I use an existing etcd cluster instead of the automatic embedded servers?
Yes, it's possible to use an existing etcd cluster instead of the automatic,
elastic embedded etcd servers. To do so, simply point to the cluster with the
`--seeds` variable, the same way you would if you were seeding a new member to
an existing mgmt cluster.
The downside to this approach is that you won't benefit from the automatic
elastic nature of the embedded etcd servers, and that you're responsible if you
accidentally break your etcd cluster, or if you use an unsupported version.
### In `mgmt` you talk about events. What is this referring to?
Mgmt has two main concepts that involve "events":
1. Events in the [resource primitive](resource-guide.md).
2. Events in the [reactive language](language-guide.md).
Each resource primitive in mgmt can test (check) and set (apply) the desired
state that was requested of it. This is familiar to what is common with existing
tools such as `Puppet`, `Ansible`, `Chef`, `Terraform`, etc... In addition,
`mgmt` can also **watch** the state and detect changes. As a result, it never
has to waste time and cpu resources by polling to test and set state, leading to
a design which is algorithmically much faster than the existing generation of
tools.
To describe the set of resources to apply, mgmt describes this collection with a
language. In order to model the time component of infrastructure, we use a
special kind of language called an [FRP](https://en.wikipedia.org/wiki/Functional_reactive_programming).
This language has a built-in concept that we call "events", and which means that
we re-evaluate the relevant portions of the code whenever a value or function
has an event that tells us that it changed. The `R` in `FRP` stands for
reactive. This is similar to how a spreadsheet updates dependent cells when a
pre-requisite value is modified. [This article](https://en.wikipedia.org/wiki/Reactive_programming)
provides a bit more background.
Whenever any of the streams of values in the language change, the program is
partially re-evaluated. The output of any mgmt program is a [DAG](https://en.wikipedia.org/wiki/Directed_acyclic_graph)
of resources, or more precisely, a stream of resource graphs. Since we have
events per-resource, we can efficiently switch from one desired-state resource
graph to the next without re-checking their individual states, since we've been
monitoring them all along.
One side-effect of all this, is that if a rogue systems administrator manually
changes the state of any managed resource, mgmt will detect this and attempt to
revert the change. This makes for excellent live demos, but is not the primary
design goal. It is a consequence of tracking state so that graph changes are
efficient. We implement the event detection via an intentional per-resource
[main loop](https://en.wikipedia.org/wiki/Event_loop) which can enable other
interesting functionality too!
Make sure to get rid of your rogue sysadmin! ;)
### Do I need to run `mgmt` as `root`?
No and yes. It depends. Nothing in mgmt explicitly requires root in the design,
however mgmt will require root only if the changes to your system that you want
it to make require root.
For example, if you use it to manage files that require root access to modify,
then you'll need root. If you only use it to manage files and resources
elsewhere, then it shouldn't need root. Many resources are perfectly usable
without root, and virtually all of my live demos are done without root.
### How can I run `mgmt` on-demand, or in `cron`, instead of continuously?
By default, `mgmt` will run continuously in an attempt to keep your machine in a
converged state, even as external forces change the current state, or as your
time-varying desired state changes over time. (You can write code in the mgmt
language which will let you describe a desired state which might change over
time.)
Some users might prefer to only run `mgmt` on-demand manually, or at a set
interval via a tool like `cron`. In order to do so, `mgmt` must have a way to
shut itself down after a single "run". This feature is possible with the
`--converged-timeout` flag. You may specify this flag, along with a number of
seconds as the argument, and when there has been no activity for that many
seconds, the program will shutdown.
Alternatively, while it is not recommended, if you'd like to ensure the program
never runs for longer that a specific number of seconds, you can ask it to
shutdown after that time interval using the `--max-runtime` flag. This also
requires a number of seconds as an argument.
#### Example:
```
./mgmt run lang examples/lang/hello0.mcl --converged-timeout=5
```
### Can I run `mgmt` for type-checking only?
Yes, you can, add the `--only-unify` option to the lang frontend while using the
`run` command, and it will exit after type unification.
#### Example:
```
./mgmt run --tmp-prefix lang --only-unify examples/lang/hello0.mcl
```
It will also print how long it took on either success or failure. Keep in mind
that even if you pass type unification, an `mgmt` run can still fail later on
for other reasons, although these are mostly runtime considerations.
### Why is type unification happening twice with `mgmt run`?
When you use the `run` action, it runs all of the compile time checks (including
type unification) that are possible, and then packages everything up into a
deployable object and runs the same mechanism that `mgmt deploy` uses, sending
the deploy to itself. At this point, mgmt starts up as a server, and receives
the deploy. It will then need to run type unification again before running the
code.
You can skip the first type unification check by adding the `--skip-unify`
option to the lang frontend when using the `run` command.
You can also skip this check when running the `deploy` action, but if your code
doesn't pass, you might be deploying broken code. This is not recommended.
#### Example:
```
./mgmt run --tmp-prefix lang --skip-unify examples/lang/hello0.mcl
```
### Why does my file resource error with `no such file or directory`?
If you create a file resource and only specify the content like this:
```
file "/tmp/foo" {
content => "hello world\n",
}
```
Then this will attempt to set the contents of that file to the desired string,
but *only* if that file already exists. If you'd like to ensure that it also
gets created in case it is not present, then you must also specify the state:
```
file "/tmp/foo" {
state => $const.res.file.state.exists,
content => "hello world\n",
}
```
Similar logic applies for situations when you only specify the `mode` parameter.
This all turns out to be more safe and "correct", in that it would error and
prevent masking an error for a situation when you expected a file to already be
at that location. It also turns out to simplify the internals significantly, and
remove an ambiguous scenario with the reversible file resource.
### Package resources error with: "The name is not activatable", what's wrong?
You may see an error like:
`main: error running auto edges: The name is not activatable`
This can happen because the mgmt `pkg` resource uses a library and daemon called
`PackageKit` to install packages. If it is not installed, then it cannot do its
work. On Fedora system you may wish to run `dnf install /usr/bin/pkcon` or on a
Debian system you may wish to run `apt install packagekit-tools`.
PackageKit is excellent because it provides both an API and an event system to
watch the package database for changes, and it abstracts away the differences
between the various package managers. If you'd prefer to not need to install
this tool, then you can contribute a native `pkg:rpm` and `pkg:deb` resource to
mgmt!
### When running mgmt, it says: "module path error: can't find a module path".
You might get an error along the lines of:
```
could not set scope: import scope `git://github.com/purpleidea/mgmt/modules/some_module_name/` failed: module path error: can't find a module path
```
This usually means that you haven't specified the directory that mgmt should use
when looking for modules. This could happen when using mgmt interactively or
when it's being run as a service. In such cases you may want the main invocation
to look something like:
```
mgmt run lang --module-path '/etc/mgmt/modules/' /etc/mgmt/main.mcl
```
### I get an error: "cannot open shared object file: No such file or directory".
Mgmt currently uses two libraries that depend on `.so` files being installed on
the host. Those are for `augeas` and `libvirt`. If those dependencies are not
present, then mgmt will not run. The complete error might look like:
```
mgmt: error while loading shared libraries: libvirt-lxc.so.0: cannot open shared object file: No such file or directory
```
or:
```
mgmt: error while loading shared libraries: libaugeas.so.0: cannot open shared object file: No such file or directory
```
or something similar. There are two solutions to this:
1. Use a build that doesn't include one or both of those features. You can build
that like: `GOTAGS="noaugeas novirt nodocker" make build`.
2. Install those dependencies. On a Fedora machine you might want to run:
```
dnf install libvirt-devel augeas-devel
```
On a Debian machine you might want to run:
```
apt install libvirt-dev libaugeas-dev
```
### Why do function names inside of templates include underscores?
The golang template library which we use to implement the golang.template() func
doesn't support the dot notation, so we import all our normal functions, and
just replace dots with underscores. As an example, the standard `datetime.print`
function is shown within mcl scripts as datetime_print after being imported.
### On startup `mgmt` hangs after: `etcd: server: starting...`.
If you get an error message similar to:
```
etcd: server: starting...
etcd: server: start timeout of 1m0s reached
etcd: server: close timeout of 15s reached
```
But nothing happens afterwards, this can be due to a corrupt etcd storage
directory. Each etcd server embedded in mgmt must have a special directory where
it stores local state. It must not be shared by more than one individual member.
This dir is typically `/var/lib/mgmt/etcd/member/`. If you accidentally use it
(for example during testing) with a different cluster view, then you can corrupt
it. This can happen if you use it with more than one different hostname.
The solution is to avoid making this mistake, and if there is no important data
saved, you can remove the etcd member dir and start over.
### On running `make` to build a new version, it errors with: `Text file busy`.
If you get an error like:
```
cp: cannot create regular file 'mgmt': Text file busy
```
This can happen if you ran `make build` (or just `make`) when there was already
an instance of mgmt running, or if a related file locking issue occurred. To
solve this, shutdown and running mgmt process, run `rm mgmt` to remove the file,
and then get a new one by running `make` again.
### Type unification error with string interpolation.
Look carefully at the following code:
```
$num = 42
print "hello" {
msg => "My favourite number is ${num}",
}
```
What's actually happening is that we can't unify `str + int`, because our
addition operator only supports `str + str`, `int + int`, and `float + float`.
The string interpolation feature can only combine strings. One solution is to
instead write:
```
$num = "42" # now this is a string
print "hello" {
msg => "My favourite number is ${num}",
}
```
The first example will usually error with something along the lines of:
`unify error with: topLevel(func() { <built-in:concat> }): type error: int != str`
Now you know why this specific case doesn't work! We may reconsider allowing
other types to be pulled into interpolation in the future. If you have a good
case for this, then let us know.
### The run and deploy commands don't parse correctly when used with `--seeds`.
If you're running a command with `--seeds`, `--server-urls`, or `--client-urls`,
then make sure you are using an equals sign between the flag name and the value.
For example, if you were to run:
```
# wrong invocation!
mgmt deploy --no-git --seeds http://127.0.0.1:2379 lang code/test.mcl
```
Then the `--seeds` flag would interpret `lang` and `code/test.mcl` as additional
seeds. This flag as well as the other aforementioned ones all accept multiple
values. Use an equals sign to guarantee you enter the correct data, eg:
```
# better invocation! (note the equals sign)
mgmt deploy --no-git --seeds=http://127.0.0.1:2379 lang code/test.mcl
```
This is caused by a parsing peculiarity of the CLI library that we are using.
This is tracked upstream at: [https://github.com/alexflint/go-arg/issues/239](https://github.com/alexflint/go-arg/issues/239).
We have a workaround in place to mitigate it and attempt to show you a helpful
error message, but it's also documented here in the meantime. The error you will
see is: `cli parse error: missing equals sign for list element`.
### The docs speaks of `--remote` but the CLI errors out?
The `--remote` flag existed in an earlier version of mgmt. It was removed and
will be replaced with a more powerful version, which is a "remote" resource. The
code is mostly ready but it's not finished. If you'd like to help finish it or
sponsor the work, please let me know.
### Does this support Windows? OSX? GNU Hurd?
Mgmt probably works best on Linux, because that's what most developers use for
serious automation workloads. Support for non-Linux operating systems isn't a
high priority of mine, but we're happy to accept patches for missing features
or resources that you think would make sense on your favourite platform.
### Why aren't you using `glide`, `godep` or `go mod` for dependency management?
Vendoring dependencies means that as the git master branch of each dependency
marches on, you're left behind using an old version. As a result, bug fixes and
improvements are not automatically brought into the project. Instead, we run our
complete test suite against the entire project (with the latest dependencies)
[every 24 hours](https://docs.travis-ci.com/user/cron-jobs/) to ensure that it
all still works.
Occasionally a dependency breaks API and causes a failure. In those situations,
we're notified almost immediately, it's easy to see exactly which commit caused
the breakage, and we can either quickly notify the author (if it was a mistake)
or update our code if it was a sensible change. This also puts less burden on
authors to support old, legacy versions of their software unnecessarily.
Historically, we've had approximately one such breakage per year, which were all
detected and fixed within a few hours. The cost of these small, rare,
interruptions is much less expensive than having to periodically move every
dependency in the project to the latest versions. Some examples of this include:
* We caught the `go-bindata` swap before it was publicly known, and fixed it in:
[adbe9c7be178898de3645b0ed17ed2ca06646017](https://github.com/purpleidea/mgmt/commit/adbe9c7be178898de3645b0ed17ed2ca06646017).
* We caught the `codegangsta/cli` API change improvement, and fixed it in:
[ab73261fd4e98cf7ecb08066ad228a8f559ba16a](https://github.com/purpleidea/mgmt/commit/ab73261fd4e98cf7ecb08066ad228a8f559ba16a).
* We caught an un-announced libvirt API change, and promptly fixed it in:
[95cb94a03958a9d2ebf01df0821a8c13a4f3a28c](https://github.com/purpleidea/mgmt/commit/95cb94a03958a9d2ebf01df0821a8c13a4f3a28c).
If we choose responsible dependencies, then it usually means that those authors
are also responsible with their changes to API and to git master. If we ever
find that it's not the case, then we will either switch that dependency to a
more responsible version, or fork it if necessary.
Occasionally, we want to pin a dependency to a particular version. This can
happen if the project treats `git master` as an unstable branch, or because a
dependency needs a newer version of golang than the minimum that we require for
our project. In those cases it's sensible to assume the technical debt, and
vendor the dependency. The common tools such as `glide` and `godep` work by
requiring you install their software, and by either storing a yaml file with the
version of that dependency in your repository, and/or copying all of that code
into git and explicitly storing it. This project thinks that all of these
solutions are wasteful and unnecessary, particularly when an existing elegant
solution already exists: `[git submodules](https://git-scm.com/book/en/v2/Git-Tools-Submodules)`.
The advantages of using `git submodules` are three-fold:
1. You already have the required tools installed.
2. You only store a pointer to the dependency, not additional files or code.
3. The git submodule tools let you easily switch dependency versions, see diff
output, and responsibly plan and test your versions bumps with ease.
Don't blindly use the tools that others tell you to. Learn what they do, think
for yourself, and become a power user today! That process led us to using
`git submodules`. Hopefully you'll come to the same conclusions that we did.
**UPDATE:**
After golang made it virtually impossible to build without `go.mod` stuff, we've
switched to it since golang 1.16. I still think the above approach was better,
and that the `go mod` tooling should have been a layer on top of git submodules
so that we don't grow yet another lock file format, and existing folks who are
comfortable with `git` can use those tools directly.
### Did you know that there is a band named `MGMT`?
I didn't realize this when naming the project, and it is accidental. After much
anguishing, I chose the name because it was short and I thought it was
appropriately descriptive. If you need a less ambiguous search term or phrase,
you can try using `mgmtconfig` or `mgmt config`.
It also doesn't stand for
[Methyl Guanine Methyl Transferase](https://en.wikipedia.org/wiki/O-6-methylguanine-DNA_methyltransferase)
which definitely existed before the band did.
### You didn't answer my question, or I have a question!
It's best to ask on [Matrix](https://matrix.to/#/#mgmtconfig:matrix.org) to see
if someone can help. If you don't get a response there, you can send a patch to
this documentation. I'll merge your question, and add a patch with the answer!
For news and updates, subscribe to the [mailing list](https://www.redhat.com/mailman/listinfo/mgmtconfig-list).

361
docs/function-guide.md Normal file
View File

@@ -0,0 +1,361 @@
# Function guide
## Overview
The `mgmt` tool has built-in functions which add useful, reactive functionality
to the language. This guide describes the different function API's that are
available. It is meant to instruct developers on how to write new functions.
Since `mgmt` and the core functions are written in golang, some prior golang
knowledge is assumed.
## Theory
Functions in `mgmt` are similar to functions in other languages, however they
also have a [reactive](https://en.wikipedia.org/wiki/Functional_reactive_programming)
component. Our functions can produce events over time, and there are different
ways to write functions. For some background on this design, please read the
[original article](https://purpleidea.com/blog/2018/02/05/mgmt-configuration-language/)
on the subject.
## Native Functions
Native functions are functions which are implemented in the mgmt language
itself. These are currently not available yet, but are coming soon. Stay tuned!
## Simple Function API
Most functions should be implemented using the simple function API. This API
allows you to implement simple, static, [pure](https://en.wikipedia.org/wiki/Pure_function)
functions that don't require you to write much boilerplate code. They will be
automatically re-evaluated as needed when their input values change. These will
all be automatically made available as helper functions within mgmt templates,
and are also available for use anywhere inside mgmt programs.
You'll need some basic knowledge of using the [`types`](https://github.com/purpleidea/mgmt/tree/master/lang/types)
library which is included with mgmt. This library lets you interact with the
available types and values in the mgmt language. It is very easy to use, and
should be fairly intuitive. Most of what you'll need to know can be inferred
from looking at example code.
To implement a function, you'll need to create a file that imports the
[`lang/funcs/simple/`](https://github.com/purpleidea/mgmt/tree/master/lang/funcs/simple/)
module. It should probably get created in the correct directory inside of:
[`lang/core/`](https://github.com/purpleidea/mgmt/tree/master/lang/core/). The
function should be implemented as a `simple.Scaffold` in our API. It is then
registered with the engine during `init()`. An example explains it best:
### Example
```golang
package simple
import (
"context"
"fmt"
"github.com/purpleidea/mgmt/lang/funcs/simple"
"github.com/purpleidea/mgmt/lang/types"
)
// you must register your functions in init when the program starts up
func init() {
// Example function that squares an int and prints out answer as an str.
simple.ModuleRegister(ModuleName, "talkingsquare", &simple.Scaffold{
T: types.NewType("func(int) str"), // declare the signature
F: func(ctx context.Context, input []types.Value) (types.Value, error) {
i := input[0].Int() // get first arg as an int64
// must return the above specified value
return &types.StrValue{
V: fmt.Sprintf("%d^2 is %d", i, i * i),
}, nil // no serious errors occurred
},
})
}
```
This simple function accepts one `int` as input, and returns one `str`.
Functions can have zero or more inputs, and must have exactly one output. You
must be sure to use the `types` library correctly, since if you try and access
an input which should not exist (eg: `input[2]`, when there are only two
that are expected), then you will cause a panic. If you have declared that a
particular argument is an `int` but you try to read it with `.Bool()` you will
also cause a panic. Lastly, make sure that you return a value in the correct
type or you will also cause a panic!
If anything goes wrong, you can return an error, however this will cause the
mgmt engine to shutdown. It should be seen as the equivalent to calling a
`panic()`, however it is safer because it brings the engine down cleanly.
Ideally, your functions should never need to error. You should never cause a
real `panic()`, since this could have negative consequences to the system.
### Example
```golang
package simple
import (
"context"
"fmt"
"github.com/purpleidea/mgmt/lang/funcs/simple"
"github.com/purpleidea/mgmt/lang/types"
)
func init() {
// This is the actual definition of the `len` function.
simple.Register("len", &simple.Scaffold{
T: types.NewType("func(?1) int"), // contains a unification var
C: simple.TypeMatch([]string{ // match on any of these sigs
"func(str) int",
"func([]?1) int",
"func(map{?1: ?2}) int",
}),
// The implementation is left as an exercise for the reader.
F: Len,
})
}
```
## Simple Polymorphic Function API
Most functions should be implemented using the simple function API. If they need
to have multiple polymorphic forms under the same name, with each resultant type
match needing to be paired to a different implementation, then you can use this
API. This is useful for situations when the functions differ in output type
only.
## Function API
To implement a reactive function in `mgmt` it must satisfy the
[`Func`](https://github.com/purpleidea/mgmt/blob/master/lang/interfaces/func.go)
interface. Using the [Simple Function API](#simple-function-api) is preferable
if it meets your needs. Most functions will be able to use that API. If you
really need something more powerful, then you can use the regular function API.
What follows are each of the method signatures and a description of each.
### Info
```golang
Info() *interfaces.Info
```
This returns some information about the function. It is necessary so that the
compiler can type check the code correctly, and know what optimizations can be
performed. This is usually the first method which is called by the engine.
#### Example
```golang
func (obj *FooFunc) Info() *interfaces.Info {
return &interfaces.Info{
Pure: true,
Sig: types.NewType("func(a int) str"),
}
}
```
### Init
```golang
Init(init *interfaces.Init) error
```
This is called to initialize the function. If something goes wrong, it should
return an error. It is passed a struct that contains all the important
information and pointers that it might need to work with throughout its
lifetime. As a result, it will need to save a copy to that pointer for future
use in the other methods.
#### Example
```golang
// Init runs some startup code for this function.
func (obj *FooFunc) Init(init *interfaces.Init) error {
obj.init = init
return nil
}
```
### Call
Call is run when you want to return a new value from the function. It takes the
input arguments to the function.
#### Example
```golang
func (obj *FooFunc) Call(ctx context.Context, args []types.Value) (types.Value, error) {
return &types.StrValue{ // Our type system "str" (string) value.
V: strconv.FormatInt(args[0].Int(), 10), // a golang string
}, nil
}
```
### Stream
```golang
Stream(context.Context) error
```
`Stream` is where any evented work is done. This method is started by the
function engine. It will run this function once. It should call the
`obj.init.Event()` method when it believes the function engine should run
`Call()` again.
Implementing this is not required if you don't have events.
If the `ctx` closes, you must shutdown as soon as possible.
#### Example
```golang
// Stream starts a mainloop and runs Event when it's time to Call() again.
func (obj *FooFunc) Stream(ctx context.Context) error {
ticker := time.NewTicker(time.Duration(1) * time.Second)
defer ticker.Stop()
// streams must generate an initial event on startup
// even though ticker will send one, we want to be faster to first event
startChan := make(chan struct{}) // start signal
close(startChan) // kick it off!
for {
select {
case <-startChan:
startChan = nil // disable
case <-ticker.C: // received the timer event
// pass
case <-ctx.Done():
return nil
}
if err := obj.init.Event(ctx); err != nil {
return err
}
}
}
```
## Further considerations
There is some additional information that any function author will need to know.
Each issue is listed separately below!
### Function struct
Each function will implement methods as pointer receivers on a function struct.
The naming convention for resources is that they end with a `Func` suffix.
#### Example
```golang
type FooFunc struct {
init *interfaces.Init
// this space can be used if needed
}
```
### Function registration
All functions must be registered with the engine so that they can be found. This
also ensures they can be encoded and decoded. Make sure to include the following
code snippet for this to work.
```golang
import "github.com/purpleidea/mgmt/lang/funcs"
func init() { // special golang method that runs once
funcs.Register("foo", func() interfaces.Func { return &FooFunc{} })
}
```
Functions inside of built-in modules will need to use the `ModuleRegister`
method instead.
```golang
// moduleName is already set to "math" by the math package. Do this in `init`.
funcs.ModuleRegister(moduleName, "cos", func() interfaces.Func {
return &CosFunc{}
})
```
### Composite functions
Composite functions are functions which import one or more existing functions.
This is useful to prevent code duplication in higher level function scenarios.
Unfortunately no further documentation about this subject has been written. To
expand this section, please send a patch! Please contact us if you'd like to
work on a function that uses this feature, or to add it to an existing one!
We don't expect this functionality to be particularly useful or common, as it's
probably easier and preferable to simply import common golang library code into
multiple different functions instead.
## Frequently asked questions
(Send your questions as a patch to this FAQ! I'll review it, merge it, and
respond by commit with the answer.)
### Can I use global variables?
Probably not. You must assume that multiple copies of your function may be used
at the same time. If they require a global variable, it's likely this won't
work. Instead it's probably better to use a struct local variable if you need to
store some state.
There might be some rare instances where a global would be acceptable, but if
you need one of these, you're probably already an internals expert. If you think
they need to lock or synchronize so as to not overwhelm an external resource,
then you have to be especially careful not to cause deadlocking the mgmt engine.
### Can I write functions in a different language?
Currently `golang` is the only supported language for built-in functions. We
might consider allowing external functions to be imported in the future. This
will likely require a language that can expose a C-like API, such as `python` or
`ruby`. Custom `golang` functions are already possible when using mgmt as a lib.
### What new functions need writing?
There are still many ideas for new functions that haven't been written yet. If
you'd like to contribute one, please contact us and tell us about your idea!
### Can I generate many different `FuncValue` implementations from one function?
Yes, you can use a function generator in `golang` to build multiple different
implementations from the same function generator. You just need to implement a
function which *returns* a `golang` type of `func([]types.Value) (types.Value, error)`
which is what `FuncValue` expects. The generator function can use any input it
wants to build the individual functions, thus helping with code reuse.
### How do I determine the signature of my simple, polymorphic function?
The determination of the input portion of the function signature can be
determined by inspecting the length of the input, and the specific type each
value has. Length is done in the standard `golang` way, and the type of each
element can be ascertained with the `Type()` method available on every value.
Knowing the output type is trickier. If it can not be inferred in some manner,
then the only way is to keep track of this yourself. You can use a function
generator to build your `FuncValue` implementations, and pass in the unique
signature to each one as you are building them. Using a generator is a common
technique which was mentioned previously.
One obvious situation where this might occur is if your function doesn't take
any inputs! An example `math.fortytwo()` function was implemented that
demonstrates the use of function generators to pass the type signatures into the
implementations.
### Where can I find more information about mgmt?
Additional blog posts, videos and other material [is available!](https://github.com/purpleidea/mgmt/blob/master/docs/on-the-web.md).
## Suggestions
If you have any ideas for API changes or other improvements to function writing,
please let us know! We're still pre 1.0 and pre 0.1 and happy to break API in
order to get it right!

795
docs/generate.go Normal file
View File

@@ -0,0 +1,795 @@
// Mgmt
// Copyright (C) James Shubin and the project contributors
// Written by James Shubin <james@shubin.ca> and the project contributors
//
// This program is free software: you can redistribute it and/or modify
// it under the terms of the GNU General Public License as published by
// the Free Software Foundation, either version 3 of the License, or
// (at your option) any later version.
//
// This program is distributed in the hope that it will be useful,
// but WITHOUT ANY WARRANTY; without even the implied warranty of
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
// GNU General Public License for more details.
//
// You should have received a copy of the GNU General Public License
// along with this program. If not, see <https://www.gnu.org/licenses/>.
//
// Additional permission under GNU GPL version 3 section 7
//
// If you modify this program, or any covered work, by linking or combining it
// with embedded mcl code and modules (and that the embedded mcl code and
// modules which link with this program, contain a copy of their source code in
// the authoritative form) containing parts covered by the terms of any other
// license, the licensors of this program grant you additional permission to
// convey the resulting work. Furthermore, the licensors of this program grant
// the original author, James Shubin, additional permission to update this
// additional permission if he deems it necessary to achieve the goals of this
// additional permission.
package docs
import (
"context"
"encoding/json"
"fmt"
"go/ast"
"go/parser"
"go/token"
"os"
"path/filepath"
"sort"
"strings"
cliUtil "github.com/purpleidea/mgmt/cli/util"
docsUtil "github.com/purpleidea/mgmt/docs/util"
"github.com/purpleidea/mgmt/engine"
engineUtil "github.com/purpleidea/mgmt/engine/util"
"github.com/purpleidea/mgmt/lang/funcs"
"github.com/purpleidea/mgmt/lang/interfaces"
"github.com/purpleidea/mgmt/util"
)
const (
// JSONSuffix is the output extension for the generated documentation.
JSONSuffix = ".json"
)
// Generate is the main entrypoint for this command. It generates everything.
type Generate struct {
*cliUtil.DocsGenerateArgs // embedded config
Config // embedded Config
// Program is the name of this program, usually set at compile time.
Program string
// Version is the version of this program, usually set at compile time.
Version string
// Debug represents if we're running in debug mode or not.
Debug bool
// Logf is a logger which should be used.
Logf func(format string, v ...interface{})
}
// Main runs everything for this setup item.
func (obj *Generate) Main(ctx context.Context) error {
if err := obj.Validate(); err != nil {
return err
}
if err := obj.Run(ctx); err != nil {
return err
}
return nil
}
// Validate verifies that the structure has acceptable data stored within.
func (obj *Generate) Validate() error {
if obj == nil {
return fmt.Errorf("data is nil")
}
if obj.Program == "" {
return fmt.Errorf("program is empty")
}
if obj.Version == "" {
return fmt.Errorf("version is empty")
}
return nil
}
// Run performs the desired actions to generate the documentation.
func (obj *Generate) Run(ctx context.Context) error {
outputFile := obj.DocsGenerateArgs.Output
if outputFile == "" || !strings.HasSuffix(outputFile, JSONSuffix) {
return fmt.Errorf("must specify output")
}
// support relative paths too!
if !strings.HasPrefix(outputFile, "/") {
wd, err := os.Getwd()
if err != nil {
return err
}
outputFile = filepath.Join(wd, outputFile)
}
if obj.Debug {
obj.Logf("output: %s", outputFile)
}
// Ensure the directory exists.
//d := filepath.Dir(outputFile)
//if err := os.MkdirAll(d, 0750); err != nil {
// return fmt.Errorf("could not make output dir at: %s", d)
//}
resources, err := obj.genResources()
if err != nil {
return err
}
functions, err := obj.genFunctions()
if err != nil {
return err
}
data := &Output{
Version: safeVersion(obj.Version),
Resources: resources,
Functions: functions,
}
b, err := json.Marshal(data)
if err != nil {
return err
}
b = append(b, '\n') // needs a trailing newline
if err := os.WriteFile(outputFile, b, 0600); err != nil {
return err
}
obj.Logf("wrote: %s", outputFile)
return nil
}
func (obj *Generate) getResourceInfo(kind, filename, structName string) (*ResourceInfo, error) {
rootDir := obj.DocsGenerateArgs.RootDir
if rootDir == "" {
wd, err := os.Getwd()
if err != nil {
return nil, err
}
rootDir = wd + "/" // add a trailing slash
}
if !strings.HasPrefix(rootDir, "/") || !strings.HasSuffix(rootDir, "/") {
return nil, fmt.Errorf("bad root dir: %s", rootDir)
}
// filename might be "noop.go" for example
p := filepath.Join(rootDir, engine.ResourcesRelDir, filename)
fset := token.NewFileSet()
// f is a: https://golang.org/pkg/go/ast/#File
f, err := parser.ParseFile(fset, p, nil, parser.ParseComments)
if err != nil {
return nil, err
}
// mcl field name to golang field name
mapping, err := engineUtil.LangFieldNameToStructFieldName(kind)
if err != nil {
return nil, err
}
// golang field name to mcl field name
nameMap, err := util.MapSwap(mapping)
if err != nil {
return nil, err
}
// mcl field name to mcl type
typMap, err := engineUtil.LangFieldNameToStructType(kind)
if err != nil {
return nil, err
}
ri := &ResourceInfo{}
// Populate the fields, even if they don't have a comment.
ri.Name = structName // golang name
ri.Kind = kind // duplicate data
ri.File = filename
ri.Fields = make(map[string]*ResourceFieldInfo)
for mclFieldName, fieldName := range mapping {
typ, exists := typMap[mclFieldName]
if !exists {
continue
}
ri.Fields[mclFieldName] = &ResourceFieldInfo{
Name: fieldName,
Type: typ.String(),
Desc: "", // empty for now
}
}
var previousComment *ast.CommentGroup
// Walk through the AST...
ast.Inspect(f, func(node ast.Node) bool {
// Comments above the struct appear as a node right _before_ we
// find the struct, so if we see one, save it for later...
if cg, ok := node.(*ast.CommentGroup); ok {
previousComment = cg
return true
}
typeSpec, ok := node.(*ast.TypeSpec)
if !ok {
return true
}
name := typeSpec.Name.Name // name is now known!
// If the struct isn't what we're expecting, then move on...
if name != structName {
return true
}
// Check if the TypeSpec is a named struct type that we want...
st, ok := typeSpec.Type.(*ast.StructType)
if !ok {
return true
}
// At this point, we have the struct we want...
var comment *ast.CommentGroup
if typeSpec.Doc != nil {
// I don't know how to even get here...
comment = typeSpec.Doc // found!
} else if previousComment != nil {
comment = previousComment // found!
previousComment = nil
}
ri.Desc = commentCleaner(comment)
// Iterate over the fields of the struct
for _, field := range st.Fields.List {
// Check if the field has a comment associated with it
if field.Doc == nil {
continue
}
if len(field.Names) < 1 { // XXX: why does this happen?
continue
}
fieldName := field.Names[0].Name
if fieldName == "" { // Can this happen?
continue
}
if isPrivate(fieldName) {
continue
}
mclFieldName, exists := nameMap[fieldName]
if !exists {
continue
}
ri.Fields[mclFieldName].Desc = commentCleaner(field.Doc)
}
return true
})
return ri, nil
}
func (obj *Generate) genResources() (map[string]*ResourceInfo, error) {
resources := make(map[string]*ResourceInfo)
if obj.DocsGenerateArgs.NoResources {
return resources, nil
}
r := engine.RegisteredResourcesNames()
sort.Strings(r)
for _, kind := range r {
metadata, err := docsUtil.LookupResource(kind)
if err != nil {
return nil, err
}
if strings.HasPrefix(kind, "_") {
// TODO: Should we display these somehow?
// built-in resource
continue
}
ri, err := obj.getResourceInfo(kind, metadata.Filename, metadata.Typename)
if err != nil {
return nil, err
}
if ri.Name == "" {
return nil, fmt.Errorf("empty resource name: %s", kind)
}
if ri.File == "" {
return nil, fmt.Errorf("empty resource file: %s", kind)
}
if ri.Desc == "" {
obj.Logf("empty resource desc: %s", kind)
}
fields := []string{}
for field := range ri.Fields {
fields = append(fields, field)
}
sort.Strings(fields)
for _, field := range fields {
if ri.Fields[field].Desc == "" {
obj.Logf("empty resource (%s) field desc: %s", kind, field)
}
}
resources[kind] = ri
}
return resources, nil
}
func (obj *Generate) getFunctionInfo(pkg, name string, metadata *docsUtil.Metadata) (*FunctionInfo, error) {
rootDir := obj.DocsGenerateArgs.RootDir
if rootDir == "" {
wd, err := os.Getwd()
if err != nil {
return nil, err
}
rootDir = wd + "/" // add a trailing slash
}
if !strings.HasPrefix(rootDir, "/") || !strings.HasSuffix(rootDir, "/") {
return nil, fmt.Errorf("bad root dir: %s", rootDir)
}
if metadata.Filename == "" {
return nil, fmt.Errorf("empty filename for: %s.%s", pkg, name)
}
// filename might be "pow.go" for example and contain a rel dir
p := filepath.Join(rootDir, funcs.FunctionsRelDir, metadata.Filename)
fset := token.NewFileSet()
// f is a: https://golang.org/pkg/go/ast/#File
f, err := parser.ParseFile(fset, p, nil, parser.ParseComments)
if err != nil {
return nil, err
}
fi := &FunctionInfo{}
fi.Name = metadata.Typename
fi.File = metadata.Filename
var previousComment *ast.CommentGroup
found := false
rawFunc := func(node ast.Node) (*ast.CommentGroup, string) {
fd, ok := node.(*ast.FuncDecl)
if !ok {
return nil, ""
}
return fd.Doc, fd.Name.Name // name is now known!
}
rawStruct := func(node ast.Node) (*ast.CommentGroup, string) {
typeSpec, ok := node.(*ast.TypeSpec)
if !ok {
return nil, ""
}
// Check if the TypeSpec is a named struct type that we want...
if _, ok := typeSpec.Type.(*ast.StructType); !ok {
return nil, ""
}
return typeSpec.Doc, typeSpec.Name.Name // name is now known!
}
// Walk through the AST...
ast.Inspect(f, func(node ast.Node) bool {
// Comments above the struct appear as a node right _before_ we
// find the struct, so if we see one, save it for later...
if cg, ok := node.(*ast.CommentGroup); ok {
previousComment = cg
return true
}
doc, name := rawFunc(node) // First see if it's a raw func.
if name == "" {
doc, name = rawStruct(node) // Otherwise it's a struct.
}
// If the func isn't what we're expecting, then move on...
if name != metadata.Typename {
return true
}
var comment *ast.CommentGroup
if doc != nil {
// I don't know how to even get here...
comment = doc // found!
} else if previousComment != nil {
comment = previousComment // found!
previousComment = nil
}
fi.Desc = commentCleaner(comment)
found = true
return true
})
if !found {
//return nil, nil
}
return fi, nil
}
func (obj *Generate) genFunctions() (map[string]*FunctionInfo, error) {
functions := make(map[string]*FunctionInfo)
if obj.DocsGenerateArgs.NoFunctions {
return functions, nil
}
m := funcs.Map() // map[string]func() interfaces.Func
names := []string{}
for name := range m {
names = append(names, name)
}
sort.Slice(names, func(i, j int) bool {
a := names[i]
b := names[j]
// TODO: do a sorted-by-package order.
return a < b
})
for _, name := range names {
//v := m[name]
//fn := v()
fn := m[name]()
// eg: golang/strings.has_suffix
sp := strings.Split(name, ".")
if len(sp) == 0 {
return nil, fmt.Errorf("unexpected empty function")
}
if len(sp) > 2 {
return nil, fmt.Errorf("unexpected function name: %s", name)
}
n := sp[0]
p := sp[0]
if len(sp) == 1 { // built-in
p = "" // no package!
}
if len(sp) == 2 { // normal import
n = sp[1]
}
if strings.HasPrefix(n, "_") {
// TODO: Should we display these somehow?
// built-in function
continue
}
var sig *string
//iface := ""
if x := fn.Info().Sig; x != nil {
s := x.String()
sig = &s
//iface = "simple"
}
metadata := &docsUtil.Metadata{}
// XXX: maybe we need a better way to get this?
mdFunc, ok := fn.(interfaces.MetadataFunc)
if !ok {
// Function doesn't tell us what the data is, let's try
// to get it automatically...
metadata.Typename = funcs.GetFunctionName(fn) // works!
metadata.Filename = "" // XXX: How can we get this?
// XXX: We only need this back-channel metadata store
// because we don't know how to get the filename without
// manually writing code in each function. Alternatively
// we could add a New() method to each struct and then
// we could modify the struct instead of having it be
// behind a copy which is needed to get new copies!
var err error
metadata, err = docsUtil.LookupFunction(name)
if err != nil {
return nil, err
}
} else if mdFunc == nil {
// programming error
return nil, fmt.Errorf("unexpected empty metadata for function: %s", name)
} else {
metadata = mdFunc.GetMetadata()
}
if metadata == nil {
return nil, fmt.Errorf("unexpected nil metadata for function: %s", name)
}
// This may be an empty func name if the function did not know
// how to get it. (This is normal for automatic regular funcs.)
if metadata.Typename == "" {
metadata.Typename = funcs.GetFunctionName(fn) // works!
}
fi, err := obj.getFunctionInfo(p, n, metadata)
if err != nil {
return nil, err
}
// We may not get any fields added if we can't find anything...
fi.Name = metadata.Typename
fi.Package = p
fi.Func = n
fi.File = metadata.Filename
//fi.Desc = desc
fi.Signature = sig
// Hack for golang generated functions!
if strings.HasPrefix(fi.Package, "golang/") && fi.File == "generated_funcs.go" {
pkg := fi.Package[len("golang/"):]
frag := strings.TrimPrefix(fi.Name, strings.Title(strings.Join(strings.Split(pkg, "/"), ""))) // yuck
fi.File = fmt.Sprintf("https://pkg.go.dev/%s#%s", pkg, frag)
}
if fi.Func == "" {
return nil, fmt.Errorf("empty function name: %s", name)
}
if fi.File == "" {
return nil, fmt.Errorf("empty function file: %s", name)
}
if fi.Desc == "" {
obj.Logf("empty function desc: %s", name)
}
if fi.Signature == nil {
obj.Logf("empty function sig: %s", name)
}
functions[name] = fi
}
return functions, nil
}
// Output is the type of the final data that will be for the json output.
type Output struct {
// Version is the sha1 or ref name of this specific version. This is
// used if we want to generate documentation with links matching the
// correct version. If unspecified then this assumes git master.
Version string `json:"version"`
// Resources contains the collection of every available resource!
// FIXME: should this be a list instead?
Resources map[string]*ResourceInfo `json:"resources"`
// Functions contains the collection of every available function!
// FIXME: should this be a list instead?
Functions map[string]*FunctionInfo `json:"functions"`
}
// ResourceInfo stores some information about each resource.
type ResourceInfo struct {
// Name is the golang name of this resource.
Name string `json:"name"`
// Kind is the kind of this resource.
Kind string `json:"kind"`
// File is the file name where this resource exists.
File string `json:"file"`
// Desc explains what this resource does.
Desc string `json:"description"`
// Fields is a collection of each resource field and corresponding info.
Fields map[string]*ResourceFieldInfo `json:"fields"`
}
// ResourceFieldInfo stores some information about each field in each resource.
type ResourceFieldInfo struct {
// Name is what this field is called in golang format.
Name string `json:"name"`
// Type is the mcl type for this field.
Type string `json:"type"`
// Desc explains what this field does.
Desc string `json:"description"`
}
// FunctionInfo stores some information about each function.
type FunctionInfo struct {
// Name is the golang name of this function. This may be an actual
// function if used by the simple API, or the name of a struct.
Name string `json:"name"`
// Package is the import name to use to get to this function.
Package string `json:"package"`
// Func is the name of the function in that package.
Func string `json:"func"`
// File is the file name where this function exists.
File string `json:"file"`
// Desc explains what this function does.
Desc string `json:"description"`
// Signature is the type signature of this function. If empty then the
// signature is not known statically and it may be polymorphic.
Signature *string `json:"signature,omitempty"`
}
// commentCleaner takes a comment group and returns it as a clean string. It
// removes the spurious newlines and programmer-focused comments. If there are
// blank lines, it replaces them with a single newline. The idea is that the
// webpage formatter would replace the newline with a <br /> or similar. This
// code is a modified alternative of the ast.CommentGroup.Text() function.
func commentCleaner(g *ast.CommentGroup) string {
if g == nil {
return ""
}
comments := make([]string, len(g.List))
for i, c := range g.List {
comments[i] = c.Text
}
lines := make([]string, 0, 10) // most comments are less than 10 lines
for _, c := range comments {
// Remove comment markers.
// The parser has given us exactly the comment text.
switch c[1] {
case '/':
//-style comment (no newline at the end)
c = c[2:]
if len(c) == 0 {
// empty line
break
}
if isDevComment(c[1:]) { // get rid of one space
continue
}
if c[0] == ' ' {
// strip first space - required for Example tests
c = c[1:]
break
}
//if isDirective(c) {
// // Ignore //go:noinline, //line, and so on.
// continue
//}
case '*':
/*-style comment */
c = c[2 : len(c)-2]
}
// Split on newlines.
cl := strings.Split(c, "\n")
// Walk lines, stripping trailing white space and adding to list.
for _, l := range cl {
lines = append(lines, stripTrailingWhitespace(l))
}
}
// Remove leading blank lines; convert runs of interior blank lines to a
// single blank line.
n := 0
for _, line := range lines {
if line != "" || n > 0 && lines[n-1] != "" {
lines[n] = line
n++
}
}
lines = lines[0:n]
// Concatenate all of these together. Blank lines should be a newline.
s := ""
for i, line := range lines {
if line == "" {
continue
}
s += line
if i < len(lines)-1 { // Is there another line?
if lines[i+1] == "" {
s += "\n" // Will eventually be a line break.
} else {
s += " "
}
}
}
return s
}
// TODO: should we use unicode.IsSpace instead?
func isWhitespace(ch byte) bool { return ch == ' ' || ch == '\t' || ch == '\n' || ch == '\r' }
// TODO: should we replace with a strings package stdlib function?
func stripTrailingWhitespace(s string) string {
i := len(s)
for i > 0 && isWhitespace(s[i-1]) {
i--
}
return s[0:i]
}
// isPrivate specifies if a field name is "private" or not.
func isPrivate(fieldName string) bool {
if fieldName == "" {
panic("invalid field name")
}
x := fieldName[0:1]
if strings.ToLower(x) == x {
return true // it was already private
}
return false
}
// isDevComment tells us that the comment is for developers only!
func isDevComment(comment string) bool {
if strings.HasPrefix(comment, "TODO:") {
return true
}
if strings.HasPrefix(comment, "FIXME:") {
return true
}
if strings.HasPrefix(comment, "XXX:") {
return true
}
return false
}
// safeVersion parses the main version string and returns a short hash for us.
// For example, we might get a string of 0.0.26-176-gabcdef012-dirty as input,
// and we'd want to return abcdef012.
func safeVersion(version string) string {
const dirty = "-dirty"
s := version
if strings.HasSuffix(s, dirty) { // helpful dirty remover
s = s[0 : len(s)-len(dirty)]
}
ix := strings.LastIndex(s, "-")
if ix == -1 { // assume we have a standalone version (future proofing?)
return s
}
s = s[ix+1:]
// From the `git describe` man page: The "g" prefix stands for "git" and
// is used to allow describing the version of a software depending on
// the SCM the software is managed with. This is useful in an
// environment where people may use different SCMs.
const g = "g"
if strings.HasPrefix(s, g) {
s = s[len(g):]
}
return s
}

17
docs/index.rst Normal file
View File

@@ -0,0 +1,17 @@
.. mgmt documentation master file, created by
sphinx-quickstart on Wed Feb 15 21:34:09 2017.
You can adapt this file completely to your liking, but it should at least
contain the root `toctree` directive.
Welcome to mgmt's documentation!
================================
.. toctree::
:maxdepth: 2
:caption: Contents:
documentation
quick-start-guide
resource-guide
prometheus
puppet-guide

1073
docs/language-guide.md Normal file

File diff suppressed because it is too large Load Diff

65
docs/on-the-web.md Normal file
View File

@@ -0,0 +1,65 @@
# On the web
Here is a list of places mgmt has appeared on the web. Feel free to send a patch
if we missed something that you think is relevant!
## Links
| Author | Format | Subject |
|---|---|---|
| James Shubin | blog | [Next generation configuration mgmt](https://purpleidea.com/blog/2016/01/18/next-generation-configuration-mgmt/) |
| James Shubin | video | [Introductory recording from DevConf.cz 2016](https://www.youtube.com/watch?v=GVhpPF0j-iE&html5=1) |
| James Shubin | video | [Introductory recording from CfgMgmtCamp.eu 2016](https://www.youtube.com/watch?v=fNeooSiIRnA&html5=1) |
| Julian Dunn | video | [On mgmt at CfgMgmtCamp.eu 2016](https://www.youtube.com/watch?v=kfF9IATUask&t=1949&html5=1) |
| Walter Heck | slides | [On mgmt at CfgMgmtCamp.eu 2016](http://www.slideshare.net/olindata/configuration-management-time-for-a-4th-generation/3) |
| Marco Marongiu | blog | [On mgmt](http://syslog.me/2016/02/15/leap-or-die/) |
| Felix Frank | blog | [From Catalog To Mgmt (on puppet to mgmt "transpiling")](https://ffrank.github.io/features/2016/02/18/from-catalog-to-mgmt/) |
| James Shubin | blog | [Automatic edges in mgmt (...and the pkg resource)](https://purpleidea.com/blog/2016/03/14/automatic-edges-in-mgmt/) |
| James Shubin | blog | [Automatic grouping in mgmt](https://purpleidea.com/blog/2016/03/30/automatic-grouping-in-mgmt/) |
| John Arundel | tweet | [“Puppets days are numbered.”](https://twitter.com/bitfield/status/732157519142002688) |
| Felix Frank | blog | [Puppet, Meet Mgmt (on puppet to mgmt internals)](https://ffrank.github.io/features/2016/06/12/puppet,-meet-mgmt/) |
| Felix Frank | blog | [Puppet Powered Mgmt (puppet to mgmt tl;dr)](https://ffrank.github.io/features/2016/06/19/puppet-powered-mgmt/) |
| James Shubin | blog | [Automatic clustering in mgmt](https://purpleidea.com/blog/2016/06/20/automatic-clustering-in-mgmt/) |
| James Shubin | video | [Recording from CoreOSFest 2016](https://www.youtube.com/watch?v=KVmDCUA42wc&html5=1) |
| James Shubin | video | [Recording from DebConf16](http://meetings-archive.debian.net/pub/debian-meetings/2016/debconf16/Next_Generation_Config_Mgmt.webm) |
| Felix Frank | blog | [Edging It All In (puppet and mgmt edges)](https://ffrank.github.io/features/2016/07/12/edging-it-all-in/) |
| Felix Frank | blog | [Translating All The Things (puppet to mgmt translation warnings)](https://ffrank.github.io/features/2016/08/19/translating-all-the-things/) |
| James Shubin | video | [Recording from systemd.conf 2016](https://www.youtube.com/watch?v=_TowsFAWWRA) |
| James Shubin | blog | [Remote execution in mgmt](https://purpleidea.com/blog/2016/10/07/remote-execution-in-mgmt/) |
| James Shubin | video | [Recording from High Load Strategy 2016](https://www.youtube.com/watch?v=-4g14KUVPVk) |
| James Shubin | video | [Recording from NLUUG 2016](https://www.youtube.com/watch?v=0vO93ni1zos) |
| James Shubin | blog | [Send/Recv in mgmt](https://purpleidea.com/blog/2016/12/07/sendrecv-in-mgmt/) |
| Julien Pivotto | blog | [Augeas resource for mgmt](https://purpleidea.com/cached/mgmt-augeas.html) (Cached from: https://roidelapluie.be/blog/2017/02/14/mgmt-augeas/) |
| James Shubin | blog | [Metaparameters in mgmt](https://purpleidea.com/blog/2017/03/01/metaparameters-in-mgmt/) |
| James Shubin | video | [Recording from Incontro DevOps 2017](https://vimeo.com/212241877) |
| Yves Brissaud | blog | [mgmt aux HumanTalks Grenoble (french)](http://log.winsos.net/2017/04/12/mgmt-aux-human-talks-grenoble.html) |
| James Shubin | video | [Recording from OSDC Berlin 2017](https://www.youtube.com/watch?v=LkEtBVLfygE&html5=1) |
| Jonathan Gold | blog | [AWS:EC2 in mgmt](https://jonathangold.ca/blog/aws-ec2-in-mgmt/) |
| James Shubin | video | [Recording from OSMC Nuremberg 2017](https://www.youtube.com/watch?v=hSVadQLeplU&html5=1) |
| James Shubin | video | [Recording from LCA 2018, Developers Miniconf](https://www.youtube.com/watch?v=OvgGfW0ilbE) |
| James Shubin | video | [Recording from LCA 2018, Sysadmin Miniconf](https://www.youtube.com/watch?v=ELq1XOJMIPY) |
| James Shubin | video | [Recording from LCA 2018, Main Conference](https://www.youtube.com/watch?v=_9PG64AOQ3w) |
| James Shubin | video | [Recording from DevConf.cz 2017](https://www.youtube.com/watch?v=-FPEK08l1Zk) |
| James Shubin | video | [Recording from FOSDEM 2018, Config Management Devroom](https://video.fosdem.org/2018/UA2.114/mgmt.webm) |
| James Shubin | blog | [Mgmt Configuration Language](https://purpleidea.com/blog/2018/02/05/mgmt-configuration-language/) |
| James Shubin | video | [Recording from CfgMgmtCamp.eu 2018](https://www.youtube.com/watch?v=NxObmwZDyrI) |
| Jonathan Gold | blog | [Go Netlink and Select](https://jonathangold.ca/blog/go-netlink-and-select/) |
| James Shubin | video | [Recording from DevOpsDays Montreal 2018](https://www.youtube.com/watch?v=1i38c5cooHo) |
| James Shubin | video | [Recording from FOSDEM Minimalistic Languages Devroom 2019](https://video.fosdem.org/2019/K.4.201/mgmtconfig.webm) |
| James Shubin | video | [Recording from FOSDEM Infra Management Devroom 2019](https://video.fosdem.org/2019/UB2.252A/mgmt.webm) |
| James Shubin | video | [Recording from FOSDEM Graph Processing Devroom 2019](https://video.fosdem.org/2019/H.1308/graph_mgmt_config.webm) |
| James Shubin | video | [Recording from FOSDEM Virtualization Devroom 2019](https://video.fosdem.org/2019/H.2213/vai_real_time_virtualization_automation.webm) |
| James Shubin | video | [Recording from FOSDEM Containers Devroom 2019](https://video.fosdem.org/2019/UA2.114/containers_mgmt.webm) |
| James Shubin | video | [Recording from FOSDEM Monitoring Devroom 2019](https://video.fosdem.org/2019/UB2.252A/real_time_merging_of_config_management_and_monitoring.webm) |
| James Shubin | blog | [Mgmt Configuration Language: Class and Include](https://purpleidea.com/blog/2019/07/26/class-and-include-in-mgmt/) |
| James Shubin | video | [Recording from FOSDEM 2020, Main Track (History)](https://video.fosdem.org/2020/Janson/automation.webm) |
| James Shubin | video | [Recording from FOSDEM 2020, Infra Management Devroom](https://video.fosdem.org/2020/UA2.120/mgmt.webm) |
| James Shubin | video | [Recording from FOSDEM 2020, Minimalistic Languages Devroom](https://video.fosdem.org/2020/AW1.125/mgmtconfigmore.webm) |
| James Shubin | video | [Recording from CfgMgmtCamp.eu 2020](https://www.youtube.com/watch?v=Kd7FAORFtsc) |
| James Shubin | video | [Recording from CfgMgmtCamp.eu 2023](https://www.youtube.com/watch?v=FeRGRj8w0BU) |
| James Shubin | video | [Recording from FOSDEM 2024, Golang Devroom](https://video.fosdem.org/2024/ud2218a/fosdem-2024-2575-single-binary-full-stack-provisioning.mp4) |
| James Shubin | video | [Recording from CfgMgmtCamp.eu 2024](https://www.youtube.com/watch?v=vBt9lpGD4bc) |
| James Shubin | blog | [Mgmt Configuration Language: Functions](https://purpleidea.com/blog/2024/11/22/functions-in-mgmt/) |
| James Shubin | blog | [Modules and imports in mgmt](https://purpleidea.com/blog/2024/12/03/modules-and-imports-in-mgmt/) |
| James Shubin | video | [Recording from FOSDEM 2025, Docs Devroom](https://video.fosdem.org/2025/k4201/fosdem-2025-6143-docs-straight-from-the-code-ast-powered-automation.mp4) |
| James Shubin | video | [Recording from CfgMgmtCamp.eu 2025](https://www.youtube.com/watch?v=0Oa7CWx4TEA) |

66
docs/prometheus.md Normal file
View File

@@ -0,0 +1,66 @@
# Prometheus support
Mgmt comes with a built-in prometheus support. It is disabled by default, and
can be enabled with the `--prometheus` command line switch.
By default, the prometheus instance will listen on [`127.0.0.1:9233`][pd]. You
can change this setting by using the `--prometheus-listen` cli option:
To have mgmt prometheus bind interface on 0.0.0.0:45001, use:
`./mgmt r --prometheus --prometheus-listen :45001`
## Metrics
Mgmt exposes three kinds of resources: _go_ metrics, _etcd_ metrics and _mgmt_
metrics.
### go metrics
We use the [prometheus go_collector][pgc] to expose go metrics. Those metrics
are mainly useful for debugging and perf testing.
### etcd metrics
mgmt exposes etcd metrics. Read more in the [upstream documentation][etcdm]
### mgmt metrics
Here is a list of the metrics we provide:
- `mgmt_resources_total`: The number of resources that mgmt is managing
- `mgmt_checkapply_total`: The number of CheckApply's that mgmt has run
- `mgmt_failures_total`: The number of resources that have failed
- `mgmt_failures`: The number of resources that have failed
- `mgmt_graph_start_time_seconds`: Start time of the current graph since unix
epoch in seconds
For each metric, you will get some extra labels:
- `kind`: The kind of mgmt resource
For `mgmt_checkapply_total`, those extra labels are set:
- `eventful`: "true" or "false", if the CheckApply triggered some changes
- `errorful`: "true" or "false", if the CheckApply reported an error
- `apply`: "true" or "false", if the CheckApply ran in apply or noop mode
## Alerting
You can use prometheus to alert you upon changes or failures. We do not provide
such templates yet, but we plan to provide some examples in this repository.
Patches welcome!
## Grafana
We do not have grafana dashboards yet. Patches welcome!
## External resources
- [prometheus website](https://prometheus.io/)
- [prometheus documentation](https://prometheus.io/docs/introduction/overview/)
- [prometheus best practices regarding metrics naming](https://prometheus.io/docs/practices/naming/)
- [grafana website](http://grafana.org/)
[pgc]: https://github.com/prometheus/client_golang/blob/master/prometheus/go_collector.go
[etcdm]: https://coreos.com/etcd/docs/latest/metrics.html
[pd]: https://github.com/prometheus/prometheus/wiki/Default-port-allocations

316
docs/puppet-guide.md Normal file
View File

@@ -0,0 +1,316 @@
# Puppet guide
`mgmt` can use Puppet as its source for the configuration graph.
This document goes into detail on how this works, and lists
some pitfalls and limitations.
For basic instructions on how to use the Puppet support, see
the [main documentation](documentation.md#puppet-support).
## Prerequisites
You need Puppet installed in your system. It is not important how you
get it. On the most common Linux distributions, you can use packages
from the OS maintainer, or upstream Puppet repositories. An alternative
that will also work on OSX is the `puppet` Ruby gem. It also has the
advantage that you can install any desired version in your home directory
or any other location.
Any release of Puppet's 3.x and 4.x series should be suitable for use with
`mgmt`. Most importantly, make sure to install the `ffrank-mgmtgraph` Puppet
module (referred to below as "the translator module").
```
puppet module install ffrank-mgmtgraph
```
Please note that the module is not required on your Puppet master (if you
use a master/agent setup). It's needed on the machine that runs `mgmt`.
You can install the module on the master anyway, so that it gets distributed
to your agents through Puppet's `pluginsync` mechanism.
### Testing the Puppet side
The following command should run successfully and print a YAML hash on your
terminal:
```puppet
puppet mgmtgraph print --code 'file { "/tmp/mgmt-test": ensure => present }'
```
You can use this CLI to test any manifests before handing them straight
to `mgmt`.
## Writing a suitable manifest
### Unsupported attributes
`mgmt` inherited its resource module from Puppet, so by and large, it's quite
possible to express `mgmt` graphs in terms of Puppet manifests. However,
there isn't (and likely never will be) full feature parity between the
respective resource types. In consequence, a manifest can have semantics that
cannot be transferred to `mgmt`.
For example, at the time of writing this, the `file` type in `mgmt` had no
notion of permissions (the file `mode`) yet. This lead to the following
warning (among others that will be discussed below):
```
$ puppet mgmtgraph print --code 'file { "/tmp/foo": mode => "0600" }'
Warning: cannot translate: File[/tmp/foo] { mode => "600" } (attribute is ignored)
```
This is a heads-up for the user, because the resulting `mgmt` graph will
in fact not pass this information to the `/tmp/foo` file resource, and
`mgmt` will ignore this file's permissions. Including such attributes in
manifests that are written expressly for `mgmt` is not sensible and should
be avoided.
### Unsupported resources
Puppet has a fairly large number of
[built-in types](https://www.puppet.com/docs/puppet/8/cheatsheet_core_types.html),
and countless more are available through
[modules](https://forge.puppet.com/). It's unlikely that all of them will
eventually receive native counterparts in `mgmt`.
When encountering an unknown resource, the translator module will replace
it with an `exec` resource in its output. This resource will run the equivalent
of a `puppet resource` command to make Puppet apply the original resource
itself. This has quite abysmal performance, because processing such a
resource requires the forking of at least one Puppet process (two if it
is found to be out of sync). This comes with considerable overhead.
On most systems, starting up any Puppet command takes several seconds.
Compared to the split second that the actual work usually takes,
this overhead can amount to several orders of magnitude.
Avoid Puppet types that `mgmt` does not implement (yet).
### Avoiding common warnings
Many resource parameters in Puppet take default values. For the most part,
the translator module just ignores them. However, there are cases in which
Puppet will default to convenient behavior that `mgmt` cannot quite replicate.
For example, translating a plain `file` resource will lead to a warning message:
```
$ puppet mgmtgraph print --code 'file { "/tmp/mgmt-test": }'
Warning: File[/tmp/mgmt-test] uses the 'puppet' file bucket, which mgmt cannot
do. There will be no backup copies!
```
The reason is that per default, Puppet assumes the following parameter value
(among others)
```puppet
file { "/tmp/mgmt-test":
backup => 'puppet',
}
```
To avoid this, specify the parameter explicitly:
```bash
puppet mgmtgraph print --code 'file { "/tmp/mgmt-test": backup => false }'
```
This is tedious in a more complex manifest. A good simplification is the
following [resource default](https://www.puppet.com/docs/puppet/8/lang_defaults)
anywhere on the top scope of your manifest:
```puppet
File { backup => false }
```
If you encounter similar warnings from other types and/or parameters,
use the same approach to silence them if possible.
## Configuring Puppet
Since `mgmt` uses an actual Puppet CLI behind the scenes, you might
need to tweak some of Puppet's runtime options in order to make it
do what you want. Reasons for this could be among the following:
* You use the `--puppet agent` variant and need to configure
`servername`, `certname` and other master/agent-related options.
* You don't want runtime information to end up in the `vardir`
that is used by your regular `puppet agent`.
* You install specific Puppet modules for `mgmt` in a non-standard
location.
`mgmt` exposes only one Puppet option in order to allow you to
control all of them, through its `--puppet-conf` option. It allows
you to specify which `puppet.conf` file should be used during
translation.
```
mgmt run puppet --puppet /opt/my-manifest.pp --puppet-conf /etc/mgmt/puppet.conf
```
Within this file, you can just specify any needed options in the
`[main]` section:
```
[main]
server=mgmt-master.example.net
vardir=/var/lib/mgmt/puppet
```
## Caveats
Please see the [README](https://github.com/ffrank/puppet-mgmtgraph/blob/master/README.md)
of the translator module for the current state of supported and unsupported
language features.
You should probably make sure to always use the latest release of
both `ffrank-mgmtgraph` and `ffrank-yamlresource` (the latter is
getting pulled in as a dependency of the former).
## Using Puppet in conjunction with the mcl lang
The graph that Puppet generates for `mgmt` can be united with a graph
that is created from native `mgmt` code in its mcl language. This is
useful when you are in the process of replacing Puppet with mgmt. You
can translate your custom modules into mgmt's language one by one,
and let mgmt run the current mix.
Instead of the usual `--puppet-conf` flag and argv for `puppet` and `mcl` input,
you need to use alternative flags to make this work:
* `--lp-lang` to specify the mcl input
* `--lp-puppet` to specify the puppet input
* `--lp-puppet-conf` to point to the optional puppet.conf file
`mgmt` will derive a graph that contains all edges and vertices from
both inputs. You essentially get two unrelated subgraphs that run in
parallel. To form edges between these subgraphs, you have to define
special vertices that will be merged. This works through a hard-coded
naming scheme.
### Mixed graph example 1 - No merges
```mcl
# lang
file "/tmp/mgmt_dir/" { state => "present" }
file "/tmp/mgmt_dir/a" { state => "present" }
```
```puppet
# puppet
file { "/tmp/puppet_dir": ensure => "directory" }
file { "/tmp/puppet_dir/a": ensure => "file" }
```
These very simple inputs (including implicit edges from directory to
respective file) result in two subgraphs that do not relate.
```
File[/tmp/mgmt_dir/] -> File[/tmp/mgmt_dir/a]
File[/tmp/puppet_dir] -> File[/tmp/puppet_dir/a]
```
### Mixed graph example 2 - Merged vertex
In order to have merged vertices in the resulting graph, you will
need to include special resources and classes in the respective
input code.
* On the lang side, add `noop` resources with names starting in `puppet_`.
* On the Puppet side, add **empty** classes with names starting in `mgmt_`.
```mcl
# lang
noop "puppet_handover_to_mgmt" {}
file "/tmp/mgmt_dir/" { state => "present" }
file "/tmp/mgmt_dir/a" { state => "present" }
Noop["puppet_handover_to_mgmt"] -> File["/tmp/mgmt_dir/"]
```
```puppet
# puppet
class mgmt_handover_to_mgmt {}
include mgmt_handover_to_mgmt
file { "/tmp/puppet_dir": ensure => "directory" }
file { "/tmp/puppet_dir/a": ensure => "file" }
File["/tmp/puppet_dir/a"] -> Class["mgmt_handover_to_mgmt"]
```
The new `noop` resource is merged with the new class, resulting in
the following graph:
```
File[/tmp/puppet_dir] -> File[/tmp/puppet_dir/a]
|
V
Noop[handover_to_mgmt]
|
V
File[/tmp/mgmt_dir/] -> File[/tmp/mgmt_dir/a]
```
You put all your ducks in a row, and the resources from the Puppet input
run before those from the mcl input.
**Note:** The names of the `noop` and the class must be identical after the
respective prefix. The common part (here, `handover_to_mgmt`) becomes the name
of the merged resource.
## Mixed graph example 3 - Multiple merges
In most scenarios, it will not be possible to define a single handover
point like in the previous example. For example, if some Puppet resources
need to run in between two stages of native resources, you need at least
two merged vertices:
```mcl
# lang
noop "puppet_handover" {}
noop "puppet_handback" {}
file "/tmp/mgmt_dir/" { state => "present" }
file "/tmp/mgmt_dir/a" { state => "present" }
file "/tmp/mgmt_dir/puppet_subtree/state-file" { state => "present" }
File["/tmp/mgmt_dir/"] -> Noop["puppet_handover"]
Noop["puppet_handback"] -> File["/tmp/mgmt_dir/puppet_subtree/state-file"]
```
```puppet
# puppet
class mgmt_handover {}
class mgmt_handback {}
include mgmt_handover, mgmt_handback
class important_stuff {
file { "/tmp/mgmt_dir/puppet_subtree":
ensure => "directory"
}
# ...
}
Class["mgmt_handover"] -> Class["important_stuff"] -> Class["mgmt_handback"]
```
The resulting graph looks roughly like this:
```
File[/tmp/mgmt_dir/] -> File[/tmp/mgmt_dir/a]
|
V
Noop[handover] -> ( class important_stuff resources )
|
V
Noop[handback]
|
V
File[/tmp/mgmt_dir/puppet_subtree/state-file]
```
You can add arbitrary numbers of merge pairs to your code bases,
with relationships as needed. From our limited experience, code
readability suffers quite a lot from these, however. We advise
to keep these structures simple.

127
docs/quick-start-guide.md Normal file
View File

@@ -0,0 +1,127 @@
# Quick start guide
## Introduction
This guide is intended for users and developers. If you're brand new to `mgmt`,
it's probably a good idea to start by reading an
[introductory article about the engine](https://purpleidea.com/blog/2016/01/18/next-generation-configuration-mgmt/)
and an [introductory article about the language](https://purpleidea.com/blog/2018/02/05/mgmt-configuration-language/).
[There are other articles and videos available](on-the-web.md) if you'd like to
learn more or prefer different formats. Once you're familiar with the general
idea, or if you prefer a hands-on approach, please start hacking...
## Getting mgmt
You can either build `mgmt` from source, or you can download a pre-built
release. There are also some distro repositories available, but they may not be
up to date. A pre-built release is the fastest option if there's one that's
available for your platform. If you are developing or testing a new patch to
`mgmt`, or there is not a release available for your platform, then you'll have
to build your own.
### Downloading a pre-built release:
The latest releases can be found [here](https://github.com/purpleidea/mgmt/releases/).
An alternate mirror is available [here](https://dl.fedoraproject.org/pub/alt/purpleidea/mgmt/releases/).
Make sure to verify the signatures of all packages before you use them. The
signing key can be downloaded from [https://purpleidea.com/contact/#pgp-key](https://purpleidea.com/contact/#pgp-key)
to verify the release.
If you've decided to install a pre-build release, you can skip to the
[Running mgmt](#running-mgmt) section below!
### Building a release:
You'll need some dependencies, including `golang`, and some associated tools.
#### Installing golang
* You need a modern golang version installed.
* To install on rpm style systems: `sudo dnf install golang`
* To install on apt style systems: `sudo apt install golang`
* To install on macOS systems install [Homebrew](https://brew.sh)
and run: `brew install go`
* You can run `go version` to check the golang version.
* If your distro is too old, you may need to [download](https://golang.org/dl/)
a newer golang version.
#### Setting up golang
* You can skip this step, as your installation will default to using `~/go/`,
but if you do not have a `GOPATH` yet and want one in a custom location, create
one and export it:
```shell
mkdir $HOME/gopath
export GOPATH=$HOME/gopath
```
* You might also want to add the GOPATH to your `~/.bashrc` or `~/.profile`.
* For more information you can read the
[GOPATH documentation](https://golang.org/cmd/go/#hdr-GOPATH_environment_variable).
#### Getting the mgmt code and associated dependencies
* Download the `mgmt` code and switch to that directory:
```shell
git clone --recursive https://github.com/purpleidea/mgmt/ ~/mgmt/
cd ~/mgmt/
```
* Add `$GOPATH/bin` to `$PATH`
```shell
export PATH=$PATH:$GOPATH/bin
```
* Run `make deps` to install system and golang dependencies. Take a look at
`misc/make-deps.sh` if you want to see the details of what it does.
#### Building mgmt
* Now run `make` to get a freshly built `mgmt` binary. If this succeeds, you can
proceed to the [Running mgmt](#running-mgmt) section below!
### Installing a distro release
Installation of `mgmt` from distribution packages currently needs improvement.
They are not always up-to-date with git master and as such are not recommended.
At the moment we have:
* [COPR](https://copr.fedoraproject.org/coprs/purpleidea/mgmt/) (currently dead)
* [Arch](https://aur.archlinux.org/packages/mgmt/) (currently stale)
Please contribute more and help improve these! We'd especially like to see a
Debian package!
### Building from a container:
This method avoids polluting your workstation with the dependencies for the
build. Here is an example using Fedora, Podman and Buildah:
```shell
git clone --recursive https://github.com/purpleidea/mgmt/
cd mgmt
docker build -t mgmt -f docker/Dockerfile .
docker run --rm --entrypoint cat mgmt mgmt > mgmt
chmod +x mgmt
./mgmt --version
# you could now copy the mgmt binary somewhere into your $PATH
# e.g., /usr/local/bin/ to make it accessible from anywhere
```
## Running mgmt
* Run `mgmt run --tmp-prefix lang examples/lang/hello0.mcl` to try out a very
simple example! If you built it from source, you'll need to use `./mgmt` from
the project directory.
* Look in that example file that you ran to see if you can figure out what it
did! You can press `^C` to exit `mgmt`.
* Have fun hacking on our future technology and get involved to shape the
project!
## Examples
Please look in the [examples/lang/](../examples/lang/) folder for some more
examples!

51
docs/release-notes/0.0.10 Normal file
View File

@@ -0,0 +1,51 @@
I've just released version 0.0.10 of mgmt!
NEWS
57 files changed, 1991 insertions(+), 752 deletions(-)
* There's a new resource called `KV`. Short examples exist, but I
haven't yet published a whole integration showing the usefulness.
* A major race was fixed. The issue of what to do with BackPokes during
start/pause was never previously solved. I had this as an open issue on
my whiteboard for a while, and I finally got some time to work through
it. The answer wasn't that difficult, but I think it was shrouded in
some tunnel vision. See the commit messages and source comments for the
details.
* The GAPI grew four new associated World API methods: StrWatch,
StrGet, StrSet, StrDel, and the associated etcd backed implementations.
These are quite useful when combined with the KV resource.
* There are now P/V style counting semaphores available as metaparams.
This is particularly cool because the implementation is (AFAIK,
assuming no bugs) dead-lock free! This is mentioned in my recent blog
post.
* See the git log for more NEWS, and sorry for anything notable I left
out!
BUGS
* There's a `concurrent map write` bug in the semaphore implementation
which is fixed in git master. Since it was a race, it was only caught
after this release was made. I should also figure out if the sema check
should go after the BackPoke or not.
MISC
We're still looking for new contributors, and there are easy, medium
and hard issues available! You're also welcome to suggest your own!
Please join us in #mgmtconfig on Freenode IRC, or ping this list if
you'd like help getting started!
THANKS
Thanks (alphabetically) to everyone who contributed to the latest
release:
James Shubin, Julien Pivotto, Michael Borden.
We had 3 unique committers since 0.0.9, and have had 30 overall.
Happy hacking,
James

59
docs/release-notes/0.0.11 Normal file
View File

@@ -0,0 +1,59 @@
I've just released version 0.0.11 of mgmt!
NEWS
20 files changed, 579 insertions(+), 126 deletions(-)
* Added a missing mutex around the semaphore map which prevents
occasional panics
* Removed exec pollint param which is not needed because of the poll
metaparam
* Fixed a state rechecking bug in exec resource (things are faster now)
* Fixed the major annoyance of exec resources receiving main's signals.
If we would ^C the main mgmt, the child processes would receive this
too which was incorrect.
* Fixed the deadlock on resource errors. This meant that previously if
a resource failed, we would deadlock the graph from shutting down. This
was bad and I'm glad it's now fixed. Sorry about that!
* Improved the backpoke logic to not require semaphores since we used
to take the lock even when we were going to backpoke which was
unnecessary.
* Added fast pausing to the graph. This means that a ^C or a pause
transition used to wait for the whole graph to roll through, but it now
finishes after the currently running resources finish executing. Read
the commit messages for more background here including the discussion
about a possible Interrupt() addition to the resource API.
* Prometheus support has been updated!
* See the git log for more NEWS, and sorry for anything notable I left
out!
BUGS
* We're in pretty good shape now. There are some small issues, but
nothing major that I don't know about.
MISC
We're still looking for new contributors, and there are easy, medium
and hard issues available! You're also welcome to suggest your own!
Please join us in #mgmtconfig on Freenode IRC, or ping this list if
you'd like help getting started!
THANKS
Thanks (alphabetically) to everyone who contributed to the latest
release:
James Shubin, Julien Pivotto
We had 2 unique committers since 0.0.10, and have had 30 overall.
Happy hacking,
James
@purpleidea

87
docs/release-notes/0.0.12 Normal file
View File

@@ -0,0 +1,87 @@
I've just released version 0.0.12 of mgmt!
Sorry if I've been more quiet than usual, I've had to focus a lot of my
time on GlusterFS related features ($dayjob official directives) and
work there.
One goal is to try and use libmgmt to either wholly or partially re-
implement glusterd. As a result, a lot of my upstream focus has been
re-prioritized to features needed for that effort.
I wrote a PoC called gd3: https://github.com/purpleidea/gd3
(It has now bit-rotted compared to upstream mgmt, but is easy to fix.)
The initial scope looks like it will be much smaller, but hopefully
this is interesting to you too. Ping me if you'd like to help.
We desperately need your contributions if we're going to get mgmt
standalone into a MVP. To motivate you, there's some great new stuff
that landed since 0.0.11, including:
* a great new YAML parser from contributor Mildred
* a huge pgraph refactor (to remove internal deps and cycles)
* a great amount of new tests and testing
and so much more...
NEWS
76 files changed, 7549 insertions(+), 4269 deletions(-)
* The svc resource now also supports user services
* There's a fabulous new yaml parser that fixes a longstanding issue in
my original code. Try it with --yaml2. I'll remove the old one
eventually. Thanks to contributor Mildred for this great work!
* Refactored the lib/ etcd usage into the GAPI's for a cleaner main
* World API grew some new methods for libmgmt users
* pgraph refactor and cleanup, now it's a fairly clean standalone pkg
* pgraph functions to flatten/merge in subgraphs along with examples
* Giant resource refactor to hopefully make things more logical, and to
simplify the resource API. This also introduces the mgraph struct to
add the higher level graph knowledge outside of pgraph.
* A partial implementation of a "Graph" (recursive subgraph?) resource!
See the code for details, as help is wanted to finish this. This should
help elucidate what the most elegant design for the mgmt core should
be.
* Send/Recv support for the exec resource as output, stdout, and stderr
* GAPI improvements to support exit requests and fast pausing
* AutoEdge API improvements including a fix+test for a regression
* A possible fix for the possible etcd server startup race
* A fun amount of new tests all over including for gometalinter
* See the git log for more NEWS, and for anything notable I left out!
BUGS
* We fixed a bunch of stuff, and added more tests!
MISC
We're still looking for new contributors, and there are easy, medium
and hard issues available! You're also welcome to suggest your own!
Please join us in #mgmtconfig on Freenode IRC, or ping this list if
you'd like help getting started!
THANKS
Thanks (alphabetically) to everyone who contributed to the latest
release:
James Shubin, Julien Pivotto, Mildred Ki'Lya
We had 3 unique committers since 0.0.11, and have had 30 overall.
Happy hacking,
James
@purpleidea

89
docs/release-notes/0.0.13 Normal file
View File

@@ -0,0 +1,89 @@
I've just released version 0.0.13 of mgmt!
I guess this is an appropriate number for a scary October release.
We recently re-licensed to the more permissive GPL. I put a lot of
thought into it, and also wrote up a short blog post about some of the
reasoning. It's here:
https://ttboj.wordpress.com/2017/10/17/copyleft-is-dead-long-live-copyl
eft/
If you read to the end you'll also see that Red Hat is not funding my
mgmt work anymore. This is too bad, and means I'll only have small
amounts of personal time available to work on this. If you'd like to
help fund my work, or know someone who'd like to, please let me know!
Having said that, there's some great new stuff that landed since
0.0.12, including:
* new resources from new contributor Jonathan Gold (aws, group, user)
* an HCL frontend from new contributor Chris McKenzie
* polish in a number of places including in the nspawn resource
and so much more... If you'd prefer to have releases more often, then
please let me know! Lastly, the language and a number of associated
parts are on its way. I hope to push this monster patch to git master
before February.
It's also worth mentioning that we have 17 resources now! wow.
NEWS
140 files changed, 3921 insertions(+), 848 deletions(-)
* Many improvements to tests, testing and small fixes to avoid false-
failures on travis.
* golint now reports 0 issues.
* An HCL frontend if you'd prefer that to the YAML. Also a great
example of how to plug in a new frontend.
* An update to golang 1.8 as the minimum version required.
* Bump the etcd version to 3.2.6+ -- Looking forward to a 3.3 release
which should probably include some patches we upstreamed.
* Addition of new user and group resources. These also include a bunch
of automatic edges.
* Addition of an AWS resource! I've wanted this for a while, as it
demonstrates nicely how event based cloud resources can fit nicely into
our design. There's still a lot to do, and we have some suggestions for
Amazon too. If you have a contact there, please put me in touch!
* Our nspawn resource is more polished now.
* See the git log for more NEWS, and for anything notable I left out!
MISC
We're still looking for new contributors, and there are easy, medium
and hard issues available! You're also welcome to suggest your own!
Please join us in #mgmtconfig on Freenode IRC, or ping this list if
you'd like help getting started!
MENTORING
We offer mentoring for new golang/mgmt hackers who want to get
involved. This is free and friendly. You get to improve your skills,
and we get some patches in return. Ping me off-list for details.
THANKS
Thanks (alphabetically) to everyone who contributed to the latest
release:
AdnanLFC, Arthur Mello, ChrisMcKenzie, Dennis Kliban, Ismael Puerto,
James Shubin, Jonathan Gold, Juan Luis de Sousa-Valadas Castaño, Juan-
Luis de Sousa-Valadas Castaño (although I suspect the last two are the
same ;))
We had 9 unique committers since 0.0.12, and have had 38 overall.
Run 'git log 0.0.12..0.0.13' to see what has changed since 0.0.12
Happy hacking,
James
@purpleidea

81
docs/release-notes/0.0.14 Normal file
View File

@@ -0,0 +1,81 @@
I've just released version 0.0.14 of mgmt!
> 118 files changed, 2688 insertions(+), 974 deletions(-)
There's some great new stuff that landed since
0.0.13, including:
* amazon AWS EC2 resource is now in git master.
* more automatic edges from new contributor Guillaume Herail (xiu)
* a move to golang 1.8 or higher
and so much more... This will probably be the last release before the
language lands in git master. It's a pretty giant patch coming :/
NEWS
* We're > 1k stars on GitHub now. It's a silly metric, but ¯\_(ツ)_/¯
* Jonathan Gold has done a lot of hard work on the AWS EC2 resource,
and it's now in git master. There are still many things we'd like to
do, but it's a great start on what is a *monster* resource, and
hopefully it will inspire others to get involved.
In particular, it was a great learning experience (for me in
particular!) about how bad the EC2 golang API is. There are some
notable design bugs we found, and if anyone from Amazon engineering
would like to reach out to us, we'd love to provide some constructive
ideas.
* Guillaume Herail (xiu) wrote some really nice patches, and picked up
on the autoedges API very quickly. Hopefully he'll have time to work on
even more!
* Paul Morgan sent us some nice shell fixups-- many more exist in an
open PR that didn't make it into this release. Hopefully we'll get
those merged by 0.0.15
* Felix Frank did a few Puppet compiler fix ups. I think he's been
refreshing his work with new resources recently...
* We're looking for help writing Google, DigitalOcean, Rackspace, etc
resources if anyone is interested, reach out to us. Particularly if
there is support from those organizations as well.
* See the git log for more NEWS, and for anything notable I left out!
EVENTS
There are a bunch of upcoming mgmt talks and events! Stay tuned for
details in the coming email, but TL;DR: Linux Conf Australia, FOSDEM, &
CfgMgmtCamp.eu -- from three different speakers, and including a
hackathon too!
MISC
We're still looking for new contributors, and there are easy, medium
and hard issues available! You're also welcome to suggest your own!
Please join us in #mgmtconfig on Freenode IRC, or ping this list if
you'd like help getting started! Expect many new tagged #mgmtlove
issues within the month.
MENTORING
We offer mentoring for new golang/mgmt hackers who want to get
involved. This is free and friendly. You get to improve your skills,
and we get some patches in return. Ping me off-list for details.
THANKS
Thanks (alphabetically) to everyone who contributed to the latest
release:
Felix Frank, Guillaume Herail, James Shubin, Jonathan Gold,
jonathangold, Julien Pivotto, Paul Morgan, Toshaan Bharvani
We had 8 unique committers since 0.0.13, and have had 41 overall.
run 'git log 0.0.13..0.0.14' to see what has changed since 0.0.13
Happy hacking,
James
@purpleidea

116
docs/release-notes/0.0.15 Normal file
View File

@@ -0,0 +1,116 @@
I've just released version 0.0.15 of mgmt!
> 328 files changed, 29869 insertions(+), 943 deletions(-)
(Yeah, that's almost 30k+ LOC)
There's some great new stuff that landed since 0.0.14, including:
* THE LANGUAGE (mcl)
* "Deploys": a distributed way to push code into your cluster
* Scheduling (as a reactive function)
* Better testing
* a move to etcd 3.3+ and golang 1.9+
and so much more... This is a monster release. Please try out the
language and all the other new features today :)
NEWS
* We released the language. Please play around with it :) It's time to
get used to this cool new paradigm. Learn more from the...
Blog post:
https://purpleidea.com/blog/2018/02/05/mgmt-configuration-language/
Video:
https://www.youtube.com/watch?v=NxObmwZDyrI
Docs:
https://github.com/purpleidea/mgmt/blob/master/docs/language-guide.md
Function guide:
https://github.com/purpleidea/mgmt/blob/master/docs/function-guide.md
And tons of code all over git master. Check the lang/ folder.
* There is a reactive scheduler in the language. Use your imagination,
or play around with:
https://github.com/purpleidea/mgmt/blob/3ad7097c8aa7eab7f895aab9af22338
c0cf82986/lang/funcs/core/schedule_polyfunc.go#L18
* There is a "deploys" feature. It's not documented yet. You should
poke around if you're curious. Consider this an early soft release.
* There is a FS implementation to store files in a POSIX-like layer on
top of etcd. It's used by deploys. It needs more tests though :)
* The language grew two "simple" API's for implementing functions, so
that new functionality can be exposed in the mgmt language.
* The language grew two ways to specify edges between resources: either
internal to the resource, or externally as standalone edges.
* The language now supports optional resource parameters via the
"elvis" operator. This keeps things type safe and avoids needing an
"undef" or "nil" in the language. This operator also works for edge
declarations.
* New contributor Johan Bloemberg has been on fire sending patches!
He has already made some great improvements to our Makefile for
testing, and the addition of the env* functions in the language, with
much more code pending in open PR's.
* Initial debian packaging has been added. It now needs a maintainer to
build, upload, and love it :)
* We have an early emacs major mode for "mcl", our language.
* Lots of new documentation has been added. Particularly for developers
wanting to contribute to the project.
* We're looking for help writing Google, DigitalOcean, Rackspace, etc
resources if anyone is interested, reach out to us. Particularly if
there is support from those organizations as well.
* See the git log for more NEWS, and for anything notable I left out!
MISC
We're still looking for new contributors, and there are easy, medium
and hard issues available! You're also welcome to suggest your own!
Please join us in #mgmtconfig on Freenode IRC, or ping this list if
you'd like help getting started! For details please see:
https://github.com/purpleidea/mgmt/blob/master/docs/faq.md#how-do-i-con
tribute-to-the-project-if-i-dont-know-golang
Many new tagged #mgmtlove issues were tagged:
https://github.com/purpleidea/mgmt/issues?q=is%3Aissue+is%3Aopen+label%
3Amgmtlove
MENTORING
We offer mentoring for new golang/mgmt hackers who want to get
involved.
This is free and friendly. You get to improve your skills,
and we get
some patches in return. Ping me off-list for details.
THANKS
Thanks (alphabetically) to everyone who contributed to the latest
release:
Carsten Thiel, dsx, James Shubin, Joe Julian, Johan Bloemberg, Jonathan
Gold, jonathangold, karimb, Oliver Frommel, Peter Oliver, Toshaan
Bharvani, Wim
We had 12 unique committers since 0.0.14, and have had 48 overall.
run 'git log 0.0.14..0.0.15' to see what has changed since 0.0.14
Happy hacking,
James
@purpleidea

104
docs/release-notes/0.0.16 Normal file
View File

@@ -0,0 +1,104 @@
I've just released version 0.0.16 of mgmt!
> 220 files changed, 14243 insertions(+), 9491 deletions(-)
Woo...
There's some great new stuff that landed since 0.0.15, including:
* A giant engine re-write! (Makes resource writing more elegant too.)
* New resources!!
* New language features!!!
and so much more... This is a monster release. Please try out the new
features today :)
NEWS
* New resources include: net, mount and docker:container. Jonathan was
responsible for all of these. Please take them for a spin! He's looking
for a job too, and would probably be happy to get paid to work on mgmt.
* We're > 1.5k stars on GitHub now. It's a silly metric, but ¯\_(ツ)_/¯
* A giant engine refactoring/re-write was done. This cleaned up the
code significantly, and made it more elegant to write resources.
Unfortunately there is one small bug that I missed and that I haven't
fixed yet. It rarely happens except during some of our tests during
shutdown, which causes intermittent failures. It shouldn't block you
playing with mgmt.
* The language "class" and "include" statements have been added. These
are important pieces for writing reusable modules which are coming
soon. Try them out! (This comes with a bunch of tests too.)
* We have an integration testing framework. It's pretty cool, it spins
up a full mgmt cluster and runs stuff. Try it out or add some tests.
* I had fun fixing a big bug: 06ee05026b0c743d19c7d62675f8ddeabdc8dd4f
* I removed the remote execution functionality from core. I realized it
could be re-written as a resource, and it was also in the way from some
other cleanups that were more important. Half the new code is done,
ping me if this is a priority for you or you want to help.
* I also removed the HCL front-end, because mcl is usable enough to be
more fun to play with, and I wanted to refactor some code. If someone
really wants it back, let me know.
* We have some release building scripts in git master, so you can now
download pre-built (with fpm) RPM, DEB, or PACMAN packages! They're
signed too. https://github.com/purpleidea/mgmt/releases/tag/0.0.16
* We're looking for help writing Google, DigitalOcean, Rackspace, etc
resources if anyone is interested, reach out to us. Particularly if
there is support from those organizations as well.
* Many other bug fixes, changes, etc...
* See the git log for more NEWS, and for anything notable I left out!
MISC
I took a bit of a break recently to catch up on some life stuff, but I
think I'm back on track. While git master hasn't been especially busy,
there's an active feature branch at feat/import which contains some fun
stuff, with a very WIP giant patch still sitting on my machine. I hope
to finish it up soon and then do another release. That branch contains
one of the last big features before I'll really be ready to run mgmt on
my personal servers!
We're still looking for new contributors, and there are easy, medium
and hard issues available! You're also welcome to suggest your own!
Please join us in #mgmtconfig on Freenode IRC, or ping this list if
you'd like help getting started! For details please see:
https://github.com/purpleidea/mgmt/blob/master/docs/faq.md#how-do-i-con
tribute-to-the-project-if-i-dont-know-golang
Many new tagged #mgmtlove issues were tagged:
https://github.com/purpleidea/mgmt/issues?q=is%3Aissue+is%3Aopen+label%
3Amgmtlove
MENTORING
We offer mentoring for new golang/mgmt hackers who want to get
involved.
This is free and friendly. You get to improve your skills,
and we get
some patches in return. Ping me off-list for details.
THANKS
Thanks (alphabetically) to everyone who contributed to the latest
release:
Alan Jenkins, James Shubin, jesus m. rodriguez, Jonathan Gold,
jonathangold, Lauri Ojansivu, phaer
We had 7 unique committers since 0.0.15, and have had 52 overall.
run 'git log 0.0.15..0.0.16' to see what has changed since 0.0.15
Happy hacking,
James
@purpleidea

111
docs/release-notes/0.0.17 Normal file
View File

@@ -0,0 +1,111 @@
I've just released version 0.0.17 of mgmt!
> 269 files changed, 13281 insertions(+), 1633 deletions(-)
There's some very useful stuff that landed since 0.0.16, including:
* Modules and import system now exists!
* A lot of tests, fixes and a huge new test infra was added
* Merging puppet with mcl code is now possible (madness!)
* We made a small change to the cli UI
* Bump to golang 1.10 (please update your environments)
And much more...
DOWNLOAD
Prebuilt binaries are available here:
https://github.com/purpleidea/mgmt/releases/tag/0.0.17
NEWS
* One of the biggest missing features was the lack of a module/import
system. After some initial exploration on what turned out to be a dead-
end, I found what I think is a very elegant approach, which is now in
this release. Please try it out, there are docs available. I hope to
write a blog post about it soon.
(There's one additional "kind" of import that I'm considering, similar
to a macro "#include", that I might add. To be determined. Let me know
if you find anything missing as of today.)
* Felix added the first version of his mcl+puppet frontend. This should
allow you to more cleverly merge legacy puppet environments with mcl
code. It's fantastic, take a look.
* Jonathan added a systemd-timer resource. This is a great replacement
for cron.
* We changed the CLI ui to improve the determinism of the frontend
selection. Basically we changed from: `mgmt run --lang code.mcl` to:
`mgmt run lang --lang code.mcl`. Remember to put --tmp-prefix after
`run` directly where it is used.
* We made a whole bunch of cleanups to the test infra, added new test
infra for testing complex mcl modules and the import/module system, and
of course added new tests.
* You can pass a list of strings as the resource name to build that
many resources. (Looping/iteration!)
* You can specify all the metaparams and auto-* properties in mcl now.
* Native mcl code can be used to write imported core code.
* There was a bug that snuck into the pkg res. This has now been fixed.
* A small, long-time copy+pasta error bug was fixed in Exec.
* Virtually all the imports of the "log" package are at the top-level
now. This will make moving to a new logger easier in the future. I have
an innovative logger idea that I have a design for that I'll eventually
get to.
* A few workarounds for occasional test failures were added. Some
legacy code needs a cleanup, and it's not done yet. Fortunately, none
of these issues seem to occur in real-life as far as I can tell, and
are caused by closing down mgmt at weird times.
* Found a bug (now fixed) in the upstream lexer Yikes! See:
57ce3fa587897d74634c1216af67dd42252c64e5
* We're looking for help writing Amazon, Google, DigitalOcean, etc
resources if anyone is interested, reach out to us. Particularly if
there is support from those organizations as well.
* Many other bug fixes, changes, etc...
* See the git log for more NEWS, and for anything notable I left out!
MISC
We're still looking for new contributors, and there are easy, medium
and hard issues available! You're also welcome to suggest your own!
Please join us in #mgmtconfig on Freenode IRC, or ping this list if
you'd like help getting started! For details please see:
https://github.com/purpleidea/mgmt/blob/master/docs/faq.md#how-do-i-con
tribute-to-the-project-if-i-dont-know-golang
Many new tagged #mgmtlove issues were tagged:
https://github.com/purpleidea/mgmt/issues?q=is%3Aissue+is%3Aopen+label%
3Amgmtlove
MENTORING
We offer mentoring for new golang/mgmt hackers who want to get
involved. This is fun and friendly! You get to improve your skills,
and we get some patches in return. Ping me off-list for details.
THANKS
Thanks (alphabetically) to everyone who contributed to the latest
release:
Felix Frank, James Shubin, Jonathan Gold, Kevin Kuehler, Michael Lesko-
Krleza, Tom Payne, Vincent Membré
We had 7 unique committers since 0.0.16, and have had 56 overall.
run 'git log 0.0.16..0.0.17' to see what has changed since 0.0.16
Happy hacking,
James
@purpleidea

106
docs/release-notes/0.0.18 Normal file
View File

@@ -0,0 +1,106 @@
I've just released version 0.0.18 of mgmt!
> 202 files changed, 5606 insertions(+), 1880 deletions(-)
There's some great stuff that landed since 0.0.17, including:
* A re-write of the core engine algorithm
* Tests are very stable
* At least three old bugs were killed
* An infra to write tests for individual resources was added
And much more...
This adds a significant amount of polish and bug fixes to mgmt. We're
on the home stretch to MVP!
DOWNLOAD
Prebuilt binaries are available here:
https://github.com/purpleidea/mgmt/releases/tag/0.0.18
NEWS
* There was a rare race that would panic the engine on shutdown. This
only seemed to happen in CPU/system call starved environments like
travis-ci. This was due to some ignorance when writing this early part
of the code base. The algorithm has been re-written, and this also
removed the use of at least one mutex. Things should be stable now, and
also much faster. (Although perf was never an issue.)
* A hidden race/deadlock in the exec resource was found and killed.
Woo! Some new tools to help find these and other problems are in misc/
* The early converger code was re-written. I was not as knowledgeable
about golang in the very beginning, and this code needed refreshing. It
contained a rare deadlock which needed to be killed.
* Toshaan added an uptime() function.
* Julien added a method for generating some simple functions for the
language.
* Lander added two new functions.
* James added a new readfile() function, and other examples.
* The template function now allows you to use imported functions. They
use underscores instead of periods for namespace separation due to a
limitation in the template library.
* Kevin and I killed a tricky race in the SocketSet code! Woo :) Kevin
also added a great cpucount() fact!
* James gave a number of presentations at FOSDEM. Some recordings are
available: https://purpleidea.com/talks/
* We're looking for help writing Amazon, Google, DigitalOcean, etc
resources if anyone is interested, reach out to us. Particularly if
there is support from those organizations as well.
* Many other bug fixes, changes, etc...
* See the git log for more NEWS, and for anything notable I left out!
BUGS
* An unfortunate bug in the type unification code was found. This can
cause small code bases to take a lot of ram/cpu to run. This will be
prioritized in an upcoming release. Until then you'll have to avoid
fancy type unification. (Specify types you know when it has to guess.)
If efficient type unification algorithms are your specialty, please let
us know, we'd like your help!
MISC
We're still looking for new contributors, and there are easy, medium
and hard issues available! You're also welcome to suggest your own!
Please join us in #mgmtconfig on Freenode IRC, or ping this list if
you'd like help getting started! For details please see:
https://github.com/purpleidea/mgmt/blob/master/docs/faq.md#how-do-i-con
tribute-to-the-project-if-i-dont-know-golang
Many tagged #mgmtlove issues exist:
https://github.com/purpleidea/mgmt/issues?q=is%3Aissue+is%3Aopen+label%
3Amgmtlove
MENTORING
We offer mentoring for new golang/mgmt hackers who want to get
involved. This is fun and friendly! You get to improve your skills,
and we get some patches in return. Ping me off-list for details.
THANKS
Thanks (alphabetically) to everyone who contributed to the latest
release:
Felix Frank, James Shubin, Jeff Waugh, Johan Bloemberg, Julien Pivotto,
Kevin Kuehler, Lander Van den Bulcke, Toshaan Bharvani
We had 8 unique committers since 0.0.17, and have had 58 overall.
run 'git log 0.0.17..0.0.18' to see what has changed since 0.0.17
Happy hacking,
James
@purpleidea

131
docs/release-notes/0.0.19 Normal file
View File

@@ -0,0 +1,131 @@
I've just released version 0.0.19 of mgmt!
> 361 files changed, 10451 insertions(+), 3919 deletions(-)
This is a very important (and huge) release and has some important
fixes that landed since 0.0.18, including:
* A huge re-write of the elastic etcd clustering code base
* A significant improvement in the type unification algorithm
* An important import/class scoping bug was discovered and fixed
* New mcl functions and resource improvements
And much more...
What comes next is just polish, new features and small bug fixes!
DOWNLOAD
Prebuilt binaries are available here:
https://github.com/purpleidea/mgmt/releases/tag/0.0.19
NEWS
* A giant etcd re-write was completed and merged. The elastic
clustering algorithm is not perfect, however it should suffice for most
use cases, and it's always possible to point mgmt at an external etcd
cluster if you don't understand the limitations of the automatic
clustering algorithm. The important part is that the core code is much
cleaner now, so hopefully races and bugs of ignorance are gone now. :)
* I found an unfortunate bug in the type unification algorithm that
severely impacted performance for some types of code bases. This is now
fixed, and I hope we should not experience problems again! Special
thanks to Sam for talking me through the problem and understanding the
space better! Woo \o/
* An important import/class scoping bug was fixed. Thanks to Nicolas
for the bug report. We also added tests for this too!
* Nicolas also added our first os detection function. os.is_debian and
os.is_redhat are now in core. Get your favourite os added today!
* The polymorphic len function can also check str length.
* The exec resource got a big cleanup. It also learned the interrupt
trait so that long running commands can be forcefully killed if need
be.
* A fancy new test infra for testing functions over time was added.
Anytime we want to check our individual FRP functions are working as
expected, this is an easy way to add a test. This way, if we ever find
a bug, we can drop in a test with the fix. This actually helped find a
very subtle bug in readfile that nobody had experienced yet!
* File res with state => exists but no content now performs as
expected.
* Improved send/recv, since it was neglected a bit. Hopefully it ends
up being a useful primitive.
* Added a new synchronization primitive that I'm called
SubscribedSignal. I found it very useful for building some of my code,
and I hope you'll find it useful too. I'd offer it upstream to the sync
package if Google didn't force their crappy CLA nonsense on everyone
who wanted to send a patch. :/ Death by 1000 paper cuts, I guess.
* Added a match function in the new regexp core package. Try it out and
add some more functions!
* Wouter has been testing mgmt and filling all sorts of useful bug
reports. We fixed at least one related to a report, and more are
planned for the next release. Wouter also sent in one cleanup patch to
remove some dead code. Welcome to the project!
* We're looking for help writing Amazon, Google, DigitalOcean, etc
resources if anyone is interested, reach out to us. Particularly if
there is support from those organizations as well.
* Many other bug fixes, changes, etc...
* See the git log for more NEWS, and for anything notable I left out!
BUGS
* There are a few known issues, in particular with some over eager
checking done in the Validate portion of two resources, that should
actually be runtime checks in CheckApply. As a result, if you intend to
change some state during the graph execution, the resource won't see
it. These should be very easy to fix if someone is interested in
writing the patch!
TALKS
* James will be presenting at this year's OSDC in Berlin. There will be
a workshop: https://osdc.de/events/mgmt-config-workshop/
and a talk:
https://osdc.de/events/mgmt-config-the-future-of-your-autonomous-datacentre/
Sign up soon if you want to guarantee a spot, as they're limited!
MISC
We're still looking for new contributors, and there are easy, medium
and hard issues available! You're also welcome to suggest your own!
Please join us in #mgmtconfig on Freenode IRC, or ping this list if
you'd like help getting started! For details please see:
https://github.com/purpleidea/mgmt/blob/master/docs/faq.md#how-do-i-con
tribute-to-the-project-if-i-dont-know-golang
Many tagged #mgmtlove issues exist:
https://github.com/purpleidea/mgmt/issues?q=is%3Aissue+is%3Aopen+label%
3Amgmtlove
MENTORING
We offer mentoring for new golang/mgmt hackers who want to get
involved. This is fun and friendly! You get to improve your skills,
and we get some patches in return. Ping me off-list for details.
THANKS
Thanks (alphabetically) to everyone who contributed to the latest
release:
Adam Sigal, Felix Frank, James Shubin, Jonathan Gold, Michael Schubert,
Mitch Fossen, Nicolas Charles, Wouter Dullaert
We had 8 unique committers since 0.0.18, and have had 63 overall.
run 'git log 0.0.18..0.0.19' to see what has changed since 0.0.18
Happy hacking,
James
@purpleidea

145
docs/release-notes/0.0.20 Normal file
View File

@@ -0,0 +1,145 @@
I've just released version 0.0.20 of mgmt!
> 295 files changed, 8585 insertions(+), 1413 deletions(-)
This was a very challenging release but it includes many useful changes
since 0.0.19, including:
* Function values / lambdas exist and are first-class
* Over 70 new tests have been added!
* Core functions, classes, and globals can now be written in mcl code
* A new compiler "Ordering" step was added
And much more...
DOWNLOAD
Prebuilt binaries are available here:
https://github.com/purpleidea/mgmt/releases/tag/0.0.20
NEWS
* After a long and challenging road, I finally got function values and
lambdas merged. If you find any bugs, please let me know! You can now
treat functions as first-class values, and even use them as closures by
capturing variable state.
* Over 70 new tests were added, most of which test the behaviour of the
new functions.
* Core packages always allowed you to write new functions in pure
golang, but now you can implement new functions, classes, and even
variables in native mcl code! It's still desirable and perhaps
efficient in some cases to want native golang implementations, but all
good programs self-host some of their stdlib in their own language
eventually.
* A new compiler step called "Ordering" was added. It's hidden inside
the SetScope step, but now lets us generate code ordering graphs and
determine exactly what to run first.
* Light copying of Node's allows more correct and efficient function
graphs that can share common vertices and edges. For example, if two
different closures capture a variable $x, they'll both use the same
copy when running the function, since the compiler can prove if they're
identical.
* Improved the type system slightly to allow advanced type comparisons.
* The type unification algorithm was improved. Hopefully it should
solve all scenarios without needing the recursive solver which was very
slow. If you find a case that isn't speedy, please let us know!
* Added subtest listing by using -short and -v in a test. This lets you
know what's available and how to run individual sub tests easily.
* Support for the systemd STATE_DIRECTORY and xdg cache dir was added
by new contributor John! Thanks!
* New contributor Adam added a pgraph test.
* A bug in the systemd mount resource was fixed. Hopefully it should
work correctly now.
* A bug that prevented us from allowing nested system imports was
fixed. Nest away! This will pave the way for us to automatically import
most of the golang standard library by doing: import "golang/regexp" or
golang/whatever".
* Added an example showing that unicode is allowed in strings.
* Fixed a rare race in the engine.
* Added some new core functions including math.mod and datetime
improvements.
* Changed the API to remove the use of --lang. This avoids the
stuttering.
* Moved to golang 1.11 and etcd 3.3.13. The later includes a fix for an
un-catchable error scenario which we fixed in etcd.
* Improved the pgraph library significantly so that we can generate
better graphs with accurate vertices based on the vertex pointers.
* Added ArchLinux OS family detection.
* We're looking for help writing Amazon, Google, DigitalOcean, etc
resources if anyone is interested, reach out to us. Particularly if
there is support from those organizations as well.
* Many other bug fixes, changes, etc...
* See the git log for more NEWS, and for anything notable I left out!
BUGS
* There are a few known issues with some disabled code paths in the new
function value code. These issues don't *need* to be fixed, but if they
are, then we should see a slight performance increase. Happy to have
someone dig into these, and they shouldn't bother anyone.
* Some of the pre-built binaries might not work on your system. We need
to start building them with the right dependencies so that `ldd`
related things are happy. For now, please try building yourself if the
build doesn't work for you, or help improve our build system.
TALKS
I'll be giving a talk at an upcoming mini-conference in Montreal. If
you're interested in attending, please let me know.
If you'd like to give an mgmt talk somewhere, please let me know!
MISC
We're still looking for new contributors, and there are easy, medium
and hard issues available! You're also welcome to suggest your own!
Please join us in #mgmtconfig on Freenode IRC, or ping this list if
you'd like help getting started! For details please see:
https://github.com/purpleidea/mgmt/blob/master/docs/faq.md#how-do-i-con
tribute-to-the-project-if-i-dont-know-golang
Many tagged #mgmtlove issues exist:
https://github.com/purpleidea/mgmt/issues?q=is%3Aissue+is%3Aopen+label%
3Amgmtlove
MENTORING
We offer mentoring for new golang/mgmt hackers who want to get
involved. This is fun and friendly! You get to improve your skills,
and we get some patches in return. Ping me off-list for details.
THANKS
Thanks (alphabetically) to everyone who contributed to the latest
release:
Adam Sigal, Christian Rebischke, Felix Frank, James Shubin, Jan
Martens, Johan Bloemberg, John Hooks, Ward Vandewege
We had 8 unique committers since 0.0.19, and have had 67 overall.
run 'git log 0.0.19..0.0.20' to see what has changed since 0.0.19
Happy hacking,
James
@purpleidea

133
docs/release-notes/0.0.21 Normal file
View File

@@ -0,0 +1,133 @@
I've just released version 0.0.21 of mgmt!
> 178 files changed, 4351 insertions(+), 829 deletions(-)
This was a very lonely release but it includes some very useful
additions since 0.0.20, including:
* The first mgmt meme!
* Working distro packages for Fedora, Debian, Ubuntu and Arch!
* Reversible resources!
* A deploy package to let you read files from the active deploy
* Improved file resource behaviours
And much more...
DOWNLOAD
Prebuilt binaries are available here:
https://github.com/purpleidea/mgmt/releases/tag/0.0.21
NEWS
* Someone made a cool mgmt meme. Look in art/mgmt_*_meme.jpg
* Distro packages are now built properly, so they should all work now.
We have builds for Fedora-30, Fedora-29, Debian-10, Ubuntu-Bionic, and
Archlinux. If you'd like a build for a different distro/version please
let me know.
* We finally got rid of the old `Compare(...) bool` API, and moved to
`Cmp(...) error`. We'll now get more useful information from Res
compares when they differ. Thanks to new contributor Donald Bakong for
working on this. He's ramping up his golang contributions, so we expect
to see more from him in the future!
* We now have reversible resources. Basically if you create a resource
and specify the reverse metaparam, eg: `Meta:reverse => true`, then if
a resource is removed (either because a new version of code doesn't
have it anymore OR more importantly if it was inside an `if` block that
became false) then the engine will perform some "reverse" action for
it. For a file, if it was added, we'll remove it. If it was edited,
we'll undo the edit. If we added ugo+w, we'll remove that. And so on.
The engine bits are done, and as well so have the file resource bits.
It should be easy to add this for any other resource where it makes
sense. This will likely be a very powerful feature that we use a lot.
* The file resource was changed slightly so that by default the "state"
is undefined. As a result, if you want a file to be created and none is
present, you need to specify the state. Otherwise specifying "content"
will only edit a file if it already exists, and otherwise be an error.
It turns out this is actually a better behaviour, even if it's not
necessarily intuitive for puppet users. It turns out it simplifies the
code drastically and makes the reversible file resource much more
logical. It seems that Puppet and Ansible got this wrong, but Cfengine
got it right. IIRC. Do you agree? (Look at the code!)
* We now catch CR \r characters in code so that you don't wonder why
the compiler is telling you about unexpected whitespace. This should
make your life easier.
* You can now read files from within the deploy. This can be used for
templates or anything else. This was one of the last missing things
that was blocking me from writing useful mcl modules.
* Fixed a copy-pasta bug where the != operator (for strings only) was
actually doing an ==. Woops! The good news is that we've been shaking
out silly bugs because I've been using mgmt more and more. Hopefully
there aren't any woops ones like this left!
* A bunch of function, class, and include tests were added. We're
getting really well tested!
* We're looking for help writing Amazon, Google, DigitalOcean, etc
resources if anyone is interested, reach out to us. Particularly if
there is support from those organizations as well.
* Many other bug fixes, changes, etc...
* See the git log for more NEWS, and for anything notable I left out!
BUGS
* Function values as args don't work yet. This is blocking us from
implementing functions like map/reduce/filter, because they'd want to
receive a function as input. TBH, I'm a bit tunnel-visioned on this,
because I'm not the compilers genius that you are. If you can help,
please let me know. I'll be posting a bunch of test cases that show
what's needed shortly.
* Three patches have been submitted to mkosi to support the image
building I've been doing. They're not merged yet, so you'll have to
apply them yourself if you want to make your own distro images. This
isn't a major requirement anyone should have, but if they're not
merged, we'll store them here and apply them as needed.
TALKS
I'll be in Belgium for FOSDEM and CfgMgmtCamp in 2020. If I'm lucky
I'll have at least one mgmt presentation. I might also consider going
to DevConf.cz if I get a talk accepted. Feel to ping me if you'd like
to hack, get consulting, training, etc while I'm in Europe!
MISC
We're still looking for new contributors, and there are easy, medium
and hard issues available! You're also welcome to suggest your own!
Please join us in #mgmtconfig on Freenode IRC, or ping this list if
you'd like help getting started! For details please see:
https://github.com/purpleidea/mgmt/blob/master/docs/faq.md#how-do-i-con
tribute-to-the-project-if-i-dont-know-golang
Many tagged #mgmtlove issues exist:
https://github.com/purpleidea/mgmt/issues?q=is%3Aissue+is%3Aopen+label%
3Amgmtlove
MENTORING
We offer mentoring for new golang/mgmt hackers who want to get
involved. This is fun and friendly! You get to improve your skills,
and we get some patches in return. Ping me off-list for details.
THANKS
Thanks (alphabetically) to everyone who contributed to the latest
release:
bjanssens, Donald Bakong, James Shubin
We had 3 unique committers since 0.0.20, and have had 69 overall.
run 'git log 0.0.20..0.0.21' to see what has changed since 0.0.20
Happy hacking,
James
@purpleidea

160
docs/release-notes/0.0.22 Normal file
View File

@@ -0,0 +1,160 @@
I've just released version 0.0.22 of mgmt!
> 579 files changed, 17984 insertions(+), 3136 deletions(-)
Compared to the last release, this is a monster. Previously:
> 178 files changed, 4351 insertions(+), 829 deletions(-)
I apologize for not doing a release earlier, but to be quite honest,
I've been busy, the people who are playing with mgmt are doing their
own builds anyways, and there are still some unimplemented, pre-
production features missing.
Also, I started a new job. News on my blog and how (if at all) it
affects mgmt is there.
This is still an incredibly important release, so let's cover some of
the points!
* This is the last release before we switch to go.mod
* IRC channel moved to #mgmtconfig on libera.chat after Freenode died
* New resources including tftp, dhcp, and http (all as servers, wow!)
* New string interpolation implementation with many tests
* Resource fields can accept complex structs and other types now
* Improved type unification solver and new invariants like generator
* A new polymorphic function API interface
And much more...
DOWNLOAD
Prebuilt binaries are NOT available here for this release:
https://github.com/purpleidea/mgmt/releases/tag/0.0.22
NEWS
* The file resource supports building files from other "fragments".
This is magic and automatic and real-time. Docs and examples are in the
repo.
* The file resource (and others) have a new trait and queryable API to
make decisions based on what other resources are in the graph.
* The file resource has a "purge" option to remove unmanaged files from
a managed directory.
* A lot of built-in functions are auto-generated from the stdlib. Most
things you would want are now present, particularly if they're pure
functions.
* There's a new consul KV resource.
* File resources support symbolic modes now!
* New tftp related resources are now present. They're great and I use
them to provision things!
* A docker image resource was added.
* First-class constants now exist. So you can do
$const.res.file.state.exists instead of typing "exists" which is prone
to typos. This is more verbose, but it's safer if that's your priority.
* We found a bug with fuzzing! Cool, thanks Patrick!
* We have dhcp server related resources. This is pretty cool when
combined with the tftp resource and you can provision a lot of stuff
from your laptop and one binary now.
* We also have http server resources. Combined with the tftp and dhcp
resources mgmt starts to look like a powerful tool to greenfield a new
datacentre and then take over and manage it continuous. All from a
single, type-safe, code base. Of course you can do other things with
this, and I'm looking forward to seeing the ideas that I haven't
thought of yet! PS: An http:ui has been partially implemented too. Ping
if you want to know more.
* Resource fields couldn't previously accept anonymous structs as types
because of how golang built its reflect library. Joe found an elegant
workaround, thanks!
* The type unification solver was improved to support some new
invariants. This makes a lot of new things possible, and was done to
support new complex functions including the eventual addition of map,
reduce, and filter. One of the new invariants is a "generator"
invariant, so that unification can take into account the entire
relevant parts of the AST. It's not a textbook CS implementation, but
it's based on sound theory I think, and it seems to work great. If you
find an edge case, please let us know.
* The polymorphic function interface was changed to use the new
unification logic. This makes a lot more sense. All the functions have
been ported to the new interface.
* 42
* We're looking for help writing Amazon, Google, DigitalOcean, etc
resources if anyone is interested, reach out to us. Particularly if
there is support from those organizations as well.
* Many other bug fixes, changes, etc...
* See the git log for more NEWS, and for anything notable I left out!
BUGS
* Function values as args don't work yet. This is blocking us from
implementing functions like map/reduce/filter, because they'd want to
receive a function as input. TBH, I'm a bit tunnel-visioned on this,
because I'm not the compilers genius that you are. If you can help,
please let me know. I'll be posting a bunch of test cases that show
what's needed shortly.
(This is the exact message I wrote last time. I've made a lot of
progress since then, but motivation here has been low. Reach out if you
can help.)
TALKS
Hopefully CfgMgmtCamp in 2022 will be back on. If we're lucky and safe,
maybe I can travel there. TBD... Feel to ping me if you'd like to hack,
talk, whatever if I'm in Europe.
MISC
We're still looking for new contributors, and there are easy, medium
and hard issues available! You're also welcome to suggest your own!
Please join us in #mgmtconfig on Libera IRC, or ping this list if
you'd like help getting started! For details please see:
https://github.com/purpleidea/mgmt/blob/master/docs/faq.md#how-do-i-con
tribute-to-the-project-if-i-dont-know-golang
Many tagged #mgmtlove issues exist:
https://github.com/purpleidea/mgmt/issues?q=is%3Aissue+is%3Aopen+label%
3Amgmtlove
Although asking in IRC is the best way to find something to work on.
MENTORING
We offer mentoring for new golang/mgmt hackers who want to get
involved. This is fun and friendly! You get to improve your skills,
and we get some patches in return. Ping me off-list for details.
THANKS
Thanks (alphabetically) to everyone who contributed to the latest
release:
Adam Sigal, Ahmed Al-Hulaibi, David Randall, Derek Buckley, Donald
Bakong, Felix Frank, Francois Rompre-Lanctot, Ivan Pejić, James Shubin,
Jean-Philippe Evrard, Jimmy Tang, Joe Groocock, Jonathan Gold, Julien
Pivotto, Kenneth Hoste, Matthew Lesko-Krleza, Patrick Meyer, viq, Yohan
Belval
We had 19 unique committers since 0.0.21, and have had 82 overall.
run 'git log 0.0.21..0.0.22' to see what has changed since 0.0.21
Happy hacking,
James
@purpleidea

102
docs/release-notes/0.0.23 Normal file
View File

@@ -0,0 +1,102 @@
I've just released version 0.0.23 of mgmt!
> 424 files changed, 7051 insertions(+), 2256 deletions(-)
This is a fairly quiet release, and I'm mostly doing this to have a
permanent tag before I start really breaking git master.
I'd like to apologize for things being kind of quiet. I've had to focus
on life and making money to pay my bills, and I've been struggling a
bit to complete some of the tougher algorithmic parts that I think are
necessary for a solid MVP. Hopefully I will succeed, but to do so, it's
going to be easier if I break git master and then sort things out
later.
I'm feeling optimistic about the future, although help with some
complex concurrent programming would certainly be appreciated.
With that, here are a few highlights from the release:
* We're using go.mod, but I'm not keeping it up-to-date regularly yet.
* We have an os.system(`cmd`) function!
* We replaced go-bindata with the embed package.
* We added a hetzner vm resource.
* We did a giant lang/ package refactor.
* Printf supports %v now (any type).
And much more...
DOWNLOAD
Prebuilt binaries are NOT available here for this release:
https://github.com/purpleidea/mgmt/releases/tag/0.0.23
NEWS
* We have our Libera IRC channel bridged to Matrix, so you can join
that way too.
* We're looking for help writing Amazon, Google, DigitalOcean, Hetzner,
etc, resources if anyone is interested, reach out to us. Particularly
if there is support from those organizations as well.
* Many other bug fixes, changes, etc...
* See the git log for more NEWS, and for anything notable I left out!
BUGS
* Function values as args don't work yet. This is blocking us from
implementing functions like map/reduce/filter, because they'd want to
receive a function as input.
(This is the exact message I wrote last time. I've made a lot of
progress since then, but motivation here has been low. Reach out if you
can help.)
TALKS
I don't have anything planned until CfgMgmtCamp 2024. If you'd like to
book me for a private event, or sponsor my travel for your conference,
please let me know.
MISC
We're still looking for new contributors, and there are easy, medium
and hard issues available! You're also welcome to suggest your own!
Please join us in #mgmtconfig on Libera IRC or Matrix, or ping this
list if you'd like help getting started! For details please see:
https://github.com/purpleidea/mgmt/blob/master/docs/faq.md#how-do-i-con
tribute-to-the-project-if-i-dont-know-golang
Many tagged #mgmtlove issues exist:
https://github.com/purpleidea/mgmt/issues?q=is%3Aissue+is%3Aopen+label%
3Amgmtlove
Although asking in IRC is the best way to find something to work on.
MENTORING
We offer mentoring for new golang/mgmt hackers who want to get
involved. This is fun and friendly! You get to improve your skills,
and we get some patches in return. Ping me off-list for details.
THANKS
Thanks (alphabetically) to everyone who contributed to the latest
release:
dantefromhell, James Shubin, Jef Masereel, Joe Groocock, Samuel
Gélineau
We had 5 unique committers since 0.0.22, and have had 85 overall.
run 'git log 0.0.22..0.0.23' to see what has changed since 0.0.22
Happy hacking,
James
@purpleidea

161
docs/release-notes/0.0.24 Normal file
View File

@@ -0,0 +1,161 @@
I've just released version 0.0.24 of mgmt!
> 600 files changed, 13622 insertions(+), 6907 deletions(-)
This is a huge and hugely important release! It has been a long time
coming. We have lambdas! I could not have done this without the
unrelentingly supportive and brilliant Samuel Gélineau.
Getting past the blockers and tricky code changes was all thanks to
him. Cleanups, polish and making it more golang idiomatic was my doing.
(The easy stuff.) If Sam wanted to be a golang expert, he could have
done it all, but it was more sensible than I do all the mundane and
filler stuff.
With that, here are a few highlights from the release:
* We have working lambdas, including iter.map =D
* We have a new function engine!
* We have improved type unification!
* We have an improved resource engine!
* We have improved many of the internal API's.
* We have so many tests.
And much more...
DOWNLOAD
Prebuilt binaries are NOT available here for this release:
https://github.com/purpleidea/mgmt/releases/tag/0.0.24
NEWS
* Our main tests are now in the excellent .txtar format. This makes
things much easier to manage.
* The iter.map function can be named as such due to parser tricks! No
need to name it xmap anymore!
* The unification solver has been greatly improved. We can infer a lot
more about function types.
* The resources API uses the context package for closing Watch and
returning early from CheckApply. The next step would be to remove all
the resource-specific timeout code and make that a metaparam.
* A new sync primitive has been added in case you'd like to use it
somewhere. We'd love help adding an even more complex one. Look in
util/sync.go for more information.
* Sam added some beautiful type inference debugging that makes things
easier for those familiar with the standard literature.
* An important bug in standalone etcd has been fixed. While embedded
etcd and automatic clustering isn't "supported" (it's buggy) the
status-quo of using your own etcd cluster is stable, and you can even
use the embedded etcd server in standalone mode...
* This means you can run `mgmt etcd` and get the standard etcd binary
behaviour that you'd get from running `etcd` normally. This makes it
easy to use both together since you only need to transport one binary
around. (And maybe mgmt will do that for you!)
* I fixed and cleaned up some sketchy code in the resource engine. I
had been unmotivated to fix this for a while because I really wanted
lambdas first, but now that they are in, I took a good look at the
code, made some fixes, and I'm really happy with it now.
* Metaparams are appropriately stateful between graph switches now.
Retry is the easy example. Limit/Burst need to be ported if you care
about these fine details.
* A RetryReset metaparam has been added. This is another good example
of how powerful metaparams are and how much potential there is for
building systems with these and future ones.
* A bunch of internal API's have been updated. This makes it better for
function and resource writers! Some GAPI changes also got pushed
through that make things clearer for those reading code.
* We have a listlookup function. It's still missing syntactic sugar
though!
* Our new function graph engine is called "dage". I think it's pretty
clever. There's a chance there is still a bug inside, but it's unclear.
Please report any issues. If you have some large machines I can test
very large and fast graphs on, please let me know.
* Lambdas really work now! The txn and ref/gc code is pretty fantastic.
* Many bugs have been killed!
* We're looking for help writing Amazon, Google, DigitalOcean, Hetzner,
etc, resources if anyone is interested, reach out to us. Particularly
if there is support from those organizations as well.
* Many other bug fixes, changes, etc...
* See the git log for more NEWS, and for anything notable I left out!
BUGS/TODO
* Function values getting _passed_ to resources doesn't work yet, but
it's not a blocker, but it would definitely be useful. We're looking
into it.
* Function graphs are unnecessarily dynamic. We might make them more
static so that we don't need as many transactions. This is really a
compiler optimization and not a bug, but it's something important we'd
like to have.
* Running two Txn's during the same pause would be really helpful. I'm
not sure how much of a performance improvement we'd get from this, but
it would sure be interesting to build. If you want to build a fancy
synchronization primitive, then let us know! Again this is not a bug.
TALKS
I don't have anything planned until CfgMgmtCamp 2024. If you'd like to
book me for a private event, or sponsor my travel for your conference,
please let me know.
MISC
We're still looking for new contributors, and there are easy, medium
and hard issues available! You're also welcome to suggest your own!
Please join us in #mgmtconfig on Libera IRC or Matrix, or ping this
list if you'd like help getting started! For details please see:
https://github.com/purpleidea/mgmt/blob/master/docs/faq.md#how-do-i-con
tribute-to-the-project-if-i-dont-know-golang
Many tagged #mgmtlove issues exist:
https://github.com/purpleidea/mgmt/issues?q=is%3Aissue+is%3Aopen+label%
3Amgmtlove
Although asking in IRC/matrix is the best way to find something to work
on.
MENTORING
We offer mentoring for new golang/mgmt hackers who want to get
involved. This is fun and friendly! You get to improve your skills,
and we get some patches in return. Ping me off-list for details.
THANKS
Thanks (alphabetically) to everyone who contributed to the latest
release:
James Shubin, Kaushal, Laurent Indermuehle, Ofek Atar, Samuel Gélineau
We had 5 unique committers since 0.0.23, and have had 88 overall.
run 'git log 0.0.23..0.0.24' to see what has changed since 0.0.23
Happy hacking,
James
@purpleidea

343
docs/release-notes/0.0.25 Normal file
View File

@@ -0,0 +1,343 @@
I've just released version 0.0.25 of mgmt!
> 686 files changed, 28391 insertions(+), 6935 deletions(-)
This is the first release that I consider to be generally useful at
solving real-world problems, without needing to be an mgmt expert. It's
also the first release that includes a very real `mcl` codebase. An
accompanying blog post is also available:
https://purpleidea.com/blog/2024/03/27/a-new-provisioning-tool/
With that, here are a few highlights from the release:
* We have a new mgmt partner program. Please sign-up for early access
to these release notes, along with other special privileges. Details
at: https://bit.ly/mgmt-partner-program
* You can build self-contained mgmt binaries that contain a custom
application. An initial "provisioning tool" has been built in this way.
Please see the blog post for more details.
* Default lookup functions exist in the language, with syntactic sugar
(the || operator) , so you can get a default value if one doesn't
exist, eg: $some_struct->some_struct_field || "some_default_value".
* Resource fields can now accept interface{} (any) types.
* A panic feature now exists in the language.
* The exec resource has new `donecmd` and `creates` fields. Of note,
`creates` supports watches too!
* Send/recv now works for autogrouped resources!
* Added `include as` (for classes) to the language. Nested
(sugar/hierarchical) classes are now supported to make this more
powerful!
* Stats are printed if the function engine is waiting for too long.
* There's a new http:flag resource, and also an http:proxy resource so
that we can have caching http proxies!
* Added a firewalld resource for opening ports!
* Added a dhcp:range resource which is very powerful and has a fun API!
* Added the "embedded" and "entry" packages, for building standalone
tools. This goes perfectly with the new CLI library that we ported
everything to.
And much more...
DOWNLOAD
Prebuilt binaries are available here for this release:
https://github.com/purpleidea/mgmt/releases/tag/0.0.25
They can also be found on the Fedora mirror:
https://dl.fedoraproject.org/pub/alt/purpleidea/mgmt/releases/0.0.25/
NEWS
* We changed the logical operators in mcl to use well-known English
tokens: OR, AND, NOT. (but in lowercase of course)
* The history function has been temporarily removed from the syntactic
core. We'll add it back if we find it's useful to have sugar!
* A bunch of lexer/parser cleanups and improvements were made.
* Default lookup functions for lists, maps, and structs have been
added. These come with syntactic sugar as mentioned above. (We plan to
keep this syntax, but we're open to feedback and changes if they're
good.)
* Resources can accept the interface{} (any) type, although this should
be used sparingly.
* We added a new mcl test suite that checks resource output too!
* Added a new `value` resource. This is a special resource kind that
can be used for building some powerful state machines. Recommended for
experienced users only.
* Improved the golang function generation to allow functions that take
[]str, so now we have a bunch more functions (like join) in our stdlib
for free.
* Add some mac address formatting functions. (core/net)
* Added a panic resource and panic function into the core language.
This is useful for safely shutting down a running mcl program to
prevent something disastrous or unhandled.
* Added a `donecmd` field to the exec resource. This runs a command
after a successful CheckApply. This replaces the `&& echo foo > done`
pattern that you'd see in some code.
* Added a new internal `local` API that can be used for local machine
operations. So far, read/writing/watching values that are stored
locally.
* Added `value` functions which bridge the `value` resource via the
`local` API. To be used sparingly!
* Bumped to golang 1.20, and we'll probably move again before the next
release.
* Allow send/recv with autogrouped resources. This adds many
possibilities, in particular with the server style resources.
* Added a bunch of tests for sneaky corner cases. Some of these were
hard to write, but I think they're worth it.
* ExprBind is now monomorphic! This was a design mistake that we
introduced, but have since repaired. We now have far fewer copies
running in the function graph, and things are much more efficient. This
means lambdas can only have one type when used at two different call
sites, which is much more logical, safer, faster and memory efficient.
* Added an --only-unify option if you want to test your code but not
run it.
* Added a concat function for the common case of interpolation. This
makes type unification significantly faster.
* Eliminated some "benign" races. You might find this commit
interesting to read: bc63b7608e84f60bf9d568188814d411a0688738
* A pgraph bug was found and fixed. A test was added too! It's amazing
this was here for so long, it just shows how subtle graph
datastructures can be.
* Added `include as` (for classes) to the language which lets our
classes produce values which can then be used elsewhere. I decided this
feature would be necessary after writing a bunch of mcl. It does have
an extraneous scoping bug, but not anything that causes problems.
* Nested classes are now supported. This lets you write the equivalent
of nested classes, without actually having to nest them! This is not
inheritance, but rather a way of handling scope and passing it
downwards.
* Improved the Ordering compiler step to catch a bunch of unhandled
bugs. Sam is a genius and was able to figure out some of these using
wizardry.
* Added some convert functions to the mcl package.
* Allow edges with colons...
* ...Because we now support a new hierarchical autogrouping algorithm!
This let's us have some very powerful resources.
* ...Like http:*, dhcp:*, and so on, but we could even go deeper!
* Fixed a super sneaky bug with resource swapping. Due to how we Cmp,
this now preserves state more often, and in particular when we need it.
I'm fairly certain that some code in a WIP branch of mine was actually
blocked because of this issue. Pleased to run into it again, but now
with a fix in place!
* Added an http:flag resource. This let's a `wget` or similar call back
to the http:server to kick off an action.
* The http:flag resource supports directories now.
* Stats are printed if the function engine is waiting for too long.
This is mostly useful for developers who are building new functions and
have a bug in their midst!
* We added a --skip-unify option to prevent the double unification when
running locally. When using `mgmt run` to test locally, we type check,
and then deploy to ourselves, which then naturally type checks again.
This skips the first one, which would be unsafe generally, but is
perfectly safe when we're running a single instance.
* Added a new http:proxy resource, and then tweaked it's API, and then
added http streaming. This is an incredibly powerful resource that lets
us build a caching http proxy with a single resource. I can't wait to
see what else it gets used for. I'm using it for provisioning. It's not
performance optimized at the moment as it uses a single mutex for
everything, but this could be extended if we wanted to scale this out.
* Added a ton of measuring/timing of common operations. This confirmed
my belief that autoedges were slower than necessary. There are two ways
to improve this. We might end up doing either one or both. Autogrouping
is currently much faster than needed, so no improvements planned for
now!
* Started to clean up the internal FS API's. It would be really great
if the core golang team would add something so we could get rid of the
afero external interfaces.
* Added an "embedded" package to offer API's related to embedded mcl
programs! This lets us build standalone binaries which are powered by
mcl.
* Moved to a new CLI (go-arg) library. This has a few downsides, but
they are fixable upstream, and this vastly improved our code quality
and API's. This needed to happen, what with the mess that was
urfave/cli. Look at our diff's, they're really elegant! This let us
clean up our lib structs as well!
* Added an "entry" package to kick-off the embedded API stuff. This
uses the new CLI API's that we just built. The end-user now has a
really easy time building new tools.
* Added a bunch of util functions to aid in building certain standalone
tools. I'm willing to accept more contributions in this space if
they're sane, and related to our general mission. Please ask and then
send patches if you're unsure.
* Added a firewalld resource which makes opening up ports automatic
when we need them. Perfect for the standalone laptop use-case.
* Made type unification cancellable in case you get into a long-running
scenario and want to end early.
* Added a `creates` field to the exec resource. Very useful, and also
supports watches! This is very useful for the common uses of exec.
* Added a dhcp:range resource to offer any number of IP addresses to
devices that we don't know the mac addresses of in advance. This makes
building a provisioning tool even more ergonomic.
* Optimized the name invariants since we can usually avoid an exclusive
invariant in the common case. This roughly halved the type unification
time. More improvements coming too!
* Caught a sneaky list type that could get through type unification
when it was interpolated alone. This now enforces the string invariant
when we specify it, which is an important language design distinction.
We added tests for this of course too!
* The "log" package has been entirely refactored and is only visible in
one place at the top of the program. Nice! I have a design for a
"better logger / user interface" if we ever want to improve on this.
* Added release targets for standalone binary builds. I also improved
the Makefile release magic significantly.
* Made a lot of small "polish" improvements to various resources.
* Most interestingly, an embedded provisioner application has been
built and made available in full. Please test and share with others.
Hopefully this will encourage more interest in the project.
* We're looking for help writing Amazon, Google, DigitalOcean, Hetzner,
etc, resources if anyone is interested, reach out to us. Particularly
if there is support from those organizations as well.
* Many other bug fixes, changes, etc...
* See the git log for more NEWS, and for anything notable I left out!
BUGS/TODO
* Function values getting _passed_ to resources doesn't work yet, but
it's not a blocker, but it would definitely be useful. We're looking
into it.
* Function graphs are unnecessarily dynamic. We might make them more
static so that we don't need as many transactions. This is really a
compiler optimization and not a bug, but it's something important we'd
like to have.
* Running two Txn's during the same pause would be really helpful. I'm
not sure how much of a performance improvement we'd get from this, but
it would sure be interesting to build. If you want to build a fancy
synchronization primitive, then let us know! Again this is not a bug.
* Type unification performance can be improved drastically. I will have
to implement the fast algorithm so that we can scale to very large mcl
programs. Help is wanted if you are familiar with "unionfind" and/or
type unification.
TALKS
I don't have anything planned until CfgMgmtCamp 2025. If you'd like to
book me for a private event, or sponsor my travel for your conference,
please let me know.
I recently gave two talks: one at CfgMgmtCamp 2024, and one at FOSDEM
in the golang room. Both are available online and demonstrated an
earlier version of the provisioning tool which is fully available
today. The talks can be found here: https://purpleidea.com/talks/
PARTNER PROGRAM
We have a new mgmt partner program which gets you early access to
releases, bug fixes, support, and many other goodies. Please sign-up
today: https://bit.ly/mgmt-partner-program
MISC
Our mailing list host (Red Hat) is no longer letting non-Red Hat
employees use their infrastructure. We're looking for a new home. I've
opened a ticket with Freedesktop. If you have any sway with them or
other recommendations, please let me know:
https://gitlab.freedesktop.org/freedesktop/freedesktop/-/issues/1082
We're still looking for new contributors, and there are easy, medium
and hard issues available! You're also welcome to suggest your own!
Please join us in #mgmtconfig on Libera IRC or Matrix (preferred) and
ping us if you'd like help getting started! For details please see:
https://github.com/purpleidea/mgmt/blob/master/docs/faq.md#how-do-i-con
tribute-to-the-project-if-i-dont-know-golang
Many tagged #mgmtlove issues exist:
https://github.com/purpleidea/mgmt/issues?q=is%3Aissue+is%3Aopen+label%
3Amgmtlove
Although asking in IRC/matrix is the best way to find something to work
on.
MENTORING
We offer mentoring for new golang/mgmt hackers who want to get
involved. This is fun and friendly! You get to improve your skills,
and we get some patches in return. Ping me off-list for details.
THANKS
Thanks (alphabetically) to everyone who contributed to the latest
release:
Eng Zer Jun, James Shubin, Oliver Lowe, Samuel Gélineau
We had 4 unique committers since 0.0.24, and have had 90 overall.
run 'git log 0.0.24..0.0.25' to see what has changed since 0.0.24
Happy hacking,
James
@purpleidea

142
docs/release-notes/0.0.26 Normal file
View File

@@ -0,0 +1,142 @@
I've just released version 0.0.26 of mgmt!
> 16 files changed, 869 insertions(+), 181 deletions(-)
Hot off the heels of the recent large release (0.0.25) I've just
released an incremental update...
See more here:
https://purpleidea.com/blog/2024/03/27/a-new-provisioning-tool/
With that, here are a few highlights from the release:
* We have a new mgmt partner program. Please sign-up for early access
to these release notes, along with other special privileges. Details
at: https://bit.ly/mgmt-partner-program
* Type unification for the provisioning tool is about 40x faster.
* We fix a small bug related to the upcoming fedora 40 release.
And much more...
DOWNLOAD
Prebuilt binaries are available here for this release:
https://github.com/purpleidea/mgmt/releases/tag/0.0.26
They can also be found on the Fedora mirror:
https://dl.fedoraproject.org/pub/alt/purpleidea/mgmt/releases/0.0.26/
NEWS
* Added old release notes into git
* We now skip over unreleased Fedora versions (like "40 Beta") when
trying to automatically determine the latest stable release.
* Type unification was structurally refactored to make way for a bunch
of future improvements and generally to modernize the code.
* Added some unification optimizations and a unification flag
optimizations system to allow solvers to support special flags. One of
these new flags was used for the provisioner code with a substantial
improvement in type unification time by about 40x.
* New cli args are also available for using these flags.
* We're looking for help writing Amazon, Google, DigitalOcean, Hetzner,
etc, resources if anyone is interested, reach out to us. Particularly
if there is support from those organizations as well.
* Many other bug fixes, changes, etc...
* See the git log for more NEWS, and for anything notable I left out!
BUGS/TODO
* Function values getting _passed_ to resources doesn't work yet, but
it's not a blocker, but it would definitely be useful. We're looking
into it.
* Function graphs are unnecessarily dynamic. We might make them more
static so that we don't need as many transactions. This is really a
compiler optimization and not a bug, but it's something important we'd
like to have.
* Running two Txn's during the same pause would be really helpful. I'm
not sure how much of a performance improvement we'd get from this, but
it would sure be interesting to build. If you want to build a fancy
synchronization primitive, then let us know! Again this is not a bug.
* General type unification performance can be improved drastically. I
will have to implement the fast algorithm so that we can scale to very
large mcl programs. Help is wanted if you are familiar with "unionfind"
and/or type unification.
TALKS
I don't have anything planned until CfgMgmtCamp 2025. If you'd like to
book me for a private event, or sponsor my travel for your conference,
please let me know.
I recently gave two talks: one at CfgMgmtCamp 2024, and one at FOSDEM
in the golang room. Both are available online and demonstrated an
earlier version of the provisioning tool which is fully available
today. The talks can be found here: https://purpleidea.com/talks/
PARTNER PROGRAM
We have a new mgmt partner program which gets you early access to
releases, bug fixes, support, and many other goodies. Please sign-up
today: https://bit.ly/mgmt-partner-program
MISC
Our mailing list host (Red Hat) is no longer letting non-Red Hat
employees use their infrastructure. We're looking for a new home. I've
opened a ticket with Freedesktop. If you have any sway with them or
other recommendations, please let me know:
https://gitlab.freedesktop.org/freedesktop/freedesktop/-/issues/1082
We're still looking for new contributors, and there are easy, medium
and hard issues available! You're also welcome to suggest your own!
Please join us in #mgmtconfig on Libera IRC or Matrix (preferred) and
ping us if you'd like help getting started! For details please see:
https://github.com/purpleidea/mgmt/blob/master/docs/faq.md#how-do-i-con
tribute-to-the-project-if-i-dont-know-golang
Many tagged #mgmtlove issues exist:
https://github.com/purpleidea/mgmt/issues?q=is%3Aissue+is%3Aopen+label%
3Amgmtlove
Although asking in IRC/matrix is the best way to find something to work
on.
MENTORING
We offer mentoring for new golang/mgmt hackers who want to get
involved. This is fun and friendly! You get to improve your skills,
and we get some patches in return. Ping me off-list for details.
THANKS
Thanks (alphabetically) to everyone who contributed to the latest
release:
James Shubin
We had 1 unique committers since 0.0.25, and have had 90 overall.
Happy hacking,
James
@purpleidea

205
docs/release-notes/0.0.27 Normal file
View File

@@ -0,0 +1,205 @@
I've just released version 0.0.27 of mgmt!
> 854 files changed, 28882 insertions(+), 16049 deletions(-)
This is rather large release, as I'm not making regular releases unless there's
a specific ask. Most folks that are playing with mgmt are using `git master`.
With that, here are a few highlights from the release:
* Type unification is now extremely fast for all scenarios.
* Added a modules/ directory with shared mcl code for everyone to use. This
includes code for virtualization, cups, shorewall, dhcp, routers, and more!
* New core mgmt commands including setup, firstboot, and docs were added!
* The provisioner got lots of improvements including handoff, and iPXE support.
And much more...
DOWNLOAD
Prebuilt binaries are available here for this release:
https://github.com/purpleidea/mgmt/releases/tag/0.0.27
They can also be found on the Fedora mirror:
https://dl.fedoraproject.org/pub/alt/purpleidea/mgmt/releases/0.0.27/
NEWS
* Primary community channel is now on Matrix. IRC is deprecated until someone
wants to run a bridge for us.
* Type unification is now textbook, and blazingly (linearly) fast. The large
programs I'm writing now unify in under 200ms. Most small programs typically
unify in ~5ms.
* Resource and edge names are always lists of strings now unless they're static.
* We're up to golang 1.23 now. Older versions may still work.
* Our type system now supports unification variables like ?1, ?2 and so on.
* I fixed a bug in my contrib.sh script which omitted the Co-authored-by people!
This means Samuel Gélineau might have previously been missed in past release
notes which is tragic, since he has been by far the most important contributor
to mgmt.
* Made toposort deterministic which fixes some spurious non-determinism.
* Added the iterator filter function. (An important core primitive.)
* Cleaned up the output of many resources to make logs more useful / less noisy.
* Added constants, although I plan to change this to a `const` import package.
* Added the list and map core packages.
* Catch $ in metaparams to make the obvious bug cases easier for users to avoid.
* Consul is now behind a build tag for now, since it's non-free. We'll remove it
eventually if there isn't a suitable free replacement.
* Added mcl modules directory with a good initial set of interesting code.
* Added the the "vardir" API to our "local" package. This is a helpful primitive
which I use in almost every module that I write.
* Added a gzip resource!
* Added a tar resource!
* We moved the template() function to the golang.template namespace. This makes
it clear what kind of template it is and de-emphasizes our "love" for it as the
blessed template engine at least for now.
* Added a sysctl resource!
* Added a virt-builder resource for building images. We can now automate virtual
machines really elegantly.
* A bunch of core functions were added including stuff in net, strings, deploy,
and more!
* The local package got a neat "pool" function. There are lots of possibilities
to use this in creative ways!
* The GAPI/deploy code got more testing and we found some edge cases and patched
them. You can now deploy in all sorts of creative ways and things should work
as expected!
* Added a resource for archiving a deploy. This is deploy:tar and helps with
bootstrapping new machines.
* Found a sneaky DHCP bug and fixed it!
* Added mgmt setup and firstboot commands! This helps bootstrap things without
needing to re-implement that logic everywhere as bash too!
* Added a "docs" command for generating resources and function documentation!
* The provisioner got lots of improvements including handoff, and iPXE support.
* New mcl modules include shorewall, dhcp, cups, some meta modules, misc modules
and more!
* Added a BMC resource in case you want to automate your server hardware.
* We now allow multiple star (*) imports although it's not recommended.
* Hostname handoff is now also part of the provisioner.
* Fixed two type unification corner cases with magic struct functions.
* Added iPXE support to the provisioner.
* Added pprof support to make it easy to generate performance information.
* Added anonymous function calling. These are occasionally useful, and now the
language has them. They were fun and concise to implement!
* We're looking for help writing Amazon, Google, DigitalOcean, Hetzner, etc,
resources if anyone is interested, reach out to us. Particularly if there is
support from those organizations as well.
* Many other bug fixes, changes, etc...
* See the git log for more NEWS, and for anything notable I left out!
BUGS/TODO
* Function values getting _passed_ to resources doesn't work yet, but it's not a
blocker, but it would definitely be useful. We're looking into it.
* Function graphs are unnecessarily dynamic. We might make them more static so
that we don't need as many transactions. This is really a compiler optimization
and not a bug, but it's something important we'd like to have.
* Running two Txn's during the same pause would be really helpful. I'm not sure
how much of a performance improvement we'd get from this, but it would sure be
interesting to build. If you want to build a fancy synchronization primitive,
then let us know! Again this is not a bug.
* The arm64 version doesn't support augeas, so it was built with:
GOTAGS='noaugeas' to get the build out.
TALKS
After FOSDEM/CfgMgmtCamp 2025, I don't have anything planned until CfgMgmtCamp
2026. If you'd like to book me for a private event, or sponsor my travel for
your conference, please let me know.
PARTNER PROGRAM
Interest in the partner program has been limited to small individuals with no
real corporate backing, so its been officially discontinued for now. If you're
interested in partnering with us and receiving support, mgmt products early
access to releases, bug fixes, support, and many other goodies, please sign-up
today: https://bit.ly/mgmt-partner-program
MISC
Our mailing list host (Red Hat) is no longer letting non-Red Hat employees use
their infrastructure. We're looking for a new home. I've opened a ticket with
Freedesktop. If you have any sway with them or other recommendations, please let
me know:
https://gitlab.freedesktop.org/freedesktop/freedesktop/-/issues/1082
We're still looking for new contributors, and while there are easy, medium and
hard issues available! You're also welcome to suggest your own! Please join us
in #mgmtconfig on Libera IRC or Matrix (preferred) and ping us if you'd like
help getting started! For details please see:
https://github.com/purpleidea/mgmt/blob/master/docs/faq.md#how-do-i-contribute-to-the-project-if-i-dont-know-golang
Many tagged #mgmtlove issues exist:
https://github.com/purpleidea/mgmt/issues?q=is%3Aissue+is%3Aopen+label%3Amgmtlove
Although asking in matrix is the best way to find something to work on.
MENTORING
We offer mentoring for new golang/mgmt hackers who want to get involved. This is
fun and friendly! You get to improve your skills, and we get some patches in
return. Ping me off-list for details.
THANKS
Thanks (alphabetically) to everyone who contributed to the latest release:
Cian Yong Leow, Felix Frank, James Shubin, Joe Groocock, Julian Rüth, Omar Al-Shuha, Samuel Gélineau, xlai89
We had 8 unique committers since 0.0.26, and have had 96 overall.
Run 'git log 0.0.26..0.0.27' to see what has changed since 0.0.26
Happy hacking,
James
@purpleidea

71
docs/release-notes/0.0.9 Normal file
View File

@@ -0,0 +1,71 @@
I've just released 0.0.9!
Release tags are now signed with my GPG key.
From now on, I'm going to try and write a release email to the list
with information about each release. Here we go!
NEWS
* There were far more commits that went into this release than I was
expecting. I delayed a while because of some nasty races and deadlocks
I encountered when cleaning some cruft out of the engine.
134 files changed, 5394 insertions(+), 2168 deletions(-)
* We grew initial integration with Prometheus thanks to new contributor
`roidelapluie`.
* The file resource can now chown/chmod files! Thanks to new
contributor `mildred`.
* The virt resource can now hotplug/hotunplug cpus. This made for some
dope demos at CfgMgmtCamp. Special thanks to pkrempa for helping me
with the libvirt API.
* Native testing of all the golang code is now enabled again. I broke
this when I split mgmt up into multiple packages. (Sorry!) Golang tests
in foo_test.go should now run and be tested automatically. I'd
especially like it if someone wrote some for pgraph/pgraph.go:
GraphSync().
* Documentation has been significantly improved. We have a resource
guide (for writing new native resources) and much more. We also now
build the docs as a RTD project:
https://mgmt.readthedocs.io/en/master/
Patches welcome to make it style as pretty as GitHub markdown does.
* See the git log for more NEWS, and sorry for anything notable I left
out!
BUGS
* I hope to resolve this soon, but it shouldn't block further
development on git master for now. The issue: if you make extremely
high speed graph changes to graphs involving the virt resource, it can
eventually cause a panic. This is being tracked here:
https://github.com/libvirt/libvirt-go/issues/7 (help welcome!)
GLUSTERFS
I've started working on some stuff for Glusterd so that it can embed
mgmt as a lib, and use the resource model to simplify cluster
management. This will involve writing Gluster resources in mgmt
(volume, brick, etc...) and all sorts of other fun stuff. If you'd like
to participate, please LMK!
MISC
We're still looking for new contributors, and there are easy, medium and hard issues available! You're also welcome to suggest your own! Please join us in #mgmtconfig on Freenode IRC, or ping this list if you'd like help getting started!
THANKS
We had 12 unique committers since 0.0.8, and have had 29 overall.
Thanks (alphabetically) to everyone who contributed to the latest
release: Daniele Sluijters, Daniel P. Berrange, Felix Frank, goberghen,
James Shubin, Julien Pivotto, Kaushal M, Lars Kulseng, Mildred Ki'Lya,
Sean Jones, Steve Milner, Tom Ritserveldt.
Happy hacking,
James
@purpleidea

280
docs/release-notes/1.0.0 Normal file
View File

@@ -0,0 +1,280 @@
I've just released version 1.0.0 of mgmt!
> 614 files changed, 30199 insertions(+), 11916 deletions(-)
This is very important and large release. It's been 10 years since I first
publicly released this project, and I might as well stop confusing new users.
I'm happily using it in production for some time now, and I love writing `mcl`
every day! I am doing customer work in mgmt, and I have happy users.
With that, here are a few highlights from the release:
* There is a new function engine which is significantly faster on large graphs.
It could be improved further, but the optimizations aren't needed for now.
* The "automatic embedded etcd clustering" should be considered deprecated. You
can run with --no-magic to ensure it's off. It was buggy and we will possibly
write it with mcl anyways. Expect it to be removed soon.
* Type unification errors have context and line numbers! Many other error
scenarios have this too! This isn't perfect, and there are still some remaining
places when you don't get this information. Please help us find and expand
these.
* The function API has been overhauled which now makes writing most functions
significantly easier and simpler. They'll also use less memory. This is a
benefit of the new function engine.
* We have added *declarative* for and forkv statements to the language.
* Exported resources are merged and gorgeous! They work how I've always wanted.
You can actually see my experiment in the very first demo of mgmt, and I finally
wrote them to work with the language how I've always wanted.
* There's an http:server:ui set of resources that have been added. Check out:
https://www.youtube.com/watch?v=8vz1MMGkuik for some examples of that in action
and more!
And much more...
SPONSORS
The `mgmt` project is generously sponsored by:
m9rx corporation - https://m9rx.com/
Please reach out if you'd like to sponsor!
DOWNLOAD
Prebuilt binaries are available here for this release:
https://github.com/purpleidea/mgmt/releases/tag/1.0.0
They can also be found on the Fedora mirror:
https://dl.fedoraproject.org/pub/alt/purpleidea/mgmt/releases/1.0.0/
NEWS
* A bunch of misc mcl code has been added to modules/ for you to see.
* The user resource has been improved following feedback from cloudflare.
* Detect self-referential frags when building files that way.
* Added a new function for URL parsing.
* Type unification errors have context and line numbers!
* There's a "baddev" feature branch which gets rebased which you can use if you
don't want to install the tools to compiler the lexer/parser stuff. We do the
ugly commit for you if that's easier for development.
* We have added *declarative* for and forkv statements to the language. If you
know of a better name that "forkv" we're happy to hear it, but a small poll
didn't produce a more convincing suggestion.
* Waiting for a deploy just happens automatically with the "empty" frontend.
* Waiting to run a deploy just waits automatically until etcd is online.
* Automatic mgmt deploying after virt provisioning works with a seeds field.
* There's a global flag to skip autoedges to improve performance.
* The docker resource has been modernized and supports running on a docker host
that we're bootstrapping.
* Docker ports were built backwards and these have been corrected.
* The "world" interface has been cleaned up dramatically. This will make life
easier for someone who wants to add a new backend there. Filesystem, scheduler,
deployer, and more are all split.
* We can run our etcd connection over SSH. That's one of the new backends.
There's actually a reconnect issue, but it's an easy fix and it should likely
come out in the next release.
* We have an is_virtual function to detect where mgmt is running!
* Virtualization modules moved to qcow2 by default. It's solid.
* Improved a lot of user-facing logging so it's clearer what's happening.
* Exported resources have been implemented ... and they're glorious. They work
how I've always dreamed, and are such a breath of fresh air from the Puppet
days. There's an export/collect system. Export works by metaparam, not a special
language feature, and collect works with core functions. It runs when the
resource in the graph actually runs, as opposed to "all at once, even if you
fail" like the old days. Yay!
* fmt.printf handles more cases!
* The file resource now has a symlink param. Someone test it and find issues.
* We have an iter.range function which is helpful with `for` statements.
* We do some speculation which drastically reduces the shape of the function
graphs in a lot of constant scenarios. This also reduces the need to change the
shape, which brings a huge performance boost.
* Don't reuse fact pointers. There was a bug around those. In fact get rid of
the fact API since it's pointless really.
* There's some new stuff in the convert package.
* We added an http:server:ui resource. This is kind of a prototype, but you can
see it in action here: https://www.youtube.com/watch?v=8vz1MMGkuik
* Fix some send/recv corner cases. I wish I had more tests for this. Hint!
* There's an os.readfilewait() function in temporarily. This will go away when
we get the <|> operator.
* A WatchFiles field was added to the exec resource. Very handy.
* We have a new "line" resource. It supports "trim"-ing too.
* There are some new functions that have been added.
* The modules/ directory got some ssh key things.
* Automatic grouping logic improved, thanks to http:server:ui stuff.
* Hierarchical grouping works very reliably as far as I can tell.
* A bunch of ctx's were added all over where they never were. Legacy code!
* A bunch of network/NetworkManager/networkd and related mcl code was added. The
interfaces are really ugly, what is the correct upstream network config thing?
* We have a modinfo function.
* We built in some ulimit settings for modern machines.
* We have an mcl class for copr setup.
* We added SSH hostkey logic into our core etcd ssh connection tooling.
* The provisioner supports exec handoff. It can also handle more scenarios, eg
booting from an ipxe usb key and not installing on it.
* The provisioner supports encrypting machines with LUKS. It does this in a very
clever way to allow creation of secure passwords after first boot. Many kudos to
the systemd and other authors who built all the needed pieces for this to just
work fairly well.
* We improved a graph function from O(n) to O(1). Woops =D
* We removed the secret channels from the function graphs. This is much simpler
now!
* ExprIf and StmtIf both do the more correct thing. I guess the bigger graph was
eventually going to need to get killed. This was a good choice that I didn't
make soon enough.
* A ton of races were killed. We're building by default with the race checker.
I don't know why I didn't do this ten years ago. Performance is not so terrible
these days, and it catches so much. Woops. Good lesson to share with others.
* The language has a nil type, but don't worry, this is only for internal
plumbing, and we will NOT let it be user facing!
* The langpuppet stuff had to be removed again for now. If it's used, patch in.
* The GAPI stuff got a major cleanup. It was early code that was bad. Now it's a
lot better.
* The new function engine is the really big story. Have a look if you're an
algorithmist. We'd love to have people work on improving it further. It's most
likely glitch free now too!
* The virt resource code a big cleanup. It runs hotplug again which had rotted
due to libvirt api changes I think.
* The qemu guest agent works automatically again.
* The svc resource (one of the earliest) has been overhauled since it had old
buggy code which has now been fixed.
* We're looking for help writing Amazon, Google, DigitalOcean, Hetzner, etc,
resources if anyone is interested, reach out to us. Particularly if there is
support from those organizations as well.
* Many other bug fixes, changes, etc...
* See the git log for more NEWS, and for anything notable I left out!
BUGS/TODO
* Function values getting _passed_ to resources doesn't work yet, but it's not a
blocker, but it would definitely be useful. We're looking into it.
* The arm64 version doesn't support augeas, so it was built with:
GOTAGS='noaugeas' to get the build out.
* We don't have the <|> operator merged yet. Expect that when we do this, we'll
consider removing the || (default) operator. This is the only pending language
change that I know of, and these cases are easily caught by the compiler and can
be easily patched.
* Autoedge performance isn't great. It can easily be disabled. Most of the time
I just specify my edges, so this is really a convenience feature, but it should
be looked into when we have a chance.
* There's a subtle ssh reconnect issue which can occur. It should be easy to
fix and I have a patch in testing.
* Our wasm code input fields grew tick marks, but I think this disturbed the
buggy wasm code. If someone is an expert here, please have at it.
TALKS
After FOSDEM/CfgMgmtCamp 2026, I don't have anything planned until CfgMgmtCamp
2027. If you'd like to book me for a private event, or sponsor my travel for
your conference, please let me know.
MISC
Our mailing list host (Red Hat) is no longer letting non-Red Hat employees use
their infrastructure. We're looking for a new home. I've opened a ticket with
Freedesktop. If you have any sway with them or other recommendations, please let
me know:
https://gitlab.freedesktop.org/freedesktop/freedesktop/-/issues/1082
We're still looking for new contributors, and while there are easy, medium and
hard issues available! You're also welcome to suggest your own! Please join us
in #mgmtconfig on Libera IRC or Matrix (preferred) and ping us if you'd like
help getting started! For details please see:
https://github.com/purpleidea/mgmt/blob/master/docs/faq.md#how-do-i-contribute-to-the-project-if-i-dont-know-golang
Many tagged #mgmtlove issues exist:
https://github.com/purpleidea/mgmt/issues?q=is%3Aissue+is%3Aopen+label%3Amgmtlove
Although asking in matrix is the best way to find something to work on.
MENTORING
We offer mentoring for new golang/mgmt hackers who want to get involved. This is
fun and friendly! You get to improve your skills, and we get some patches in
return. Ping me off-list for details.
THANKS
Thanks (alphabetically) to everyone who contributed to the latest release:
Ahmad Abuziad, Edward Toroshchyn, Felix Frank, hades, James Shubin, Karpfen, Lourenço, Lourenço Vales, Samuel Gélineau
We had 10 unique committers since 0.0.27, and have had 103 overall.
run 'git log 0.0.27..1.0.0' to see what has changed since 0.0.27
Happy hacking,
James
@purpleidea

Some files were not shown because too many files have changed in this diff Show More