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!

docs/resource-guide.md

@@ -333,7 +333,6 @@ func (obj *FooRes) Watch(processChan chan *event.Event) error {
 	var send = false // send event?
 	var exit *error
 	for {
-		obj.SetState(ResStateWatching) // reset
 		select {
 		case event := <-obj.Events():
 			cuid.SetConverged(false)

pgraph/actions.go

@@ -62,16 +62,15 @@ func (g *Graph) OKTimestamp(v *Vertex) bool {
 	return true
 }
 
-// Poke notifies nodes after me in the dependency graph that they need refreshing...
-// NOTE: this assumes that this can never fail or need to be rescheduled
-func (g *Graph) Poke(v *Vertex, activity bool) error {
+// Poke tells nodes after me in the dependency graph that they need to refresh.
+func (g *Graph) Poke(v *Vertex) error {
 	var wg sync.WaitGroup
 	// these are all the vertices pointing AWAY FROM v, eg: v -> ???
 	for _, n := range g.OutgoingGraphVertices(v) {
-		// XXX: if we're in state event and haven't been cancelled by
-		// apply, then we can cancel a poke to a child, right? XXX
-		// XXX: if n.Res.getState() != resources.ResStateEvent || activity { // is this correct?
-		if true || activity { // XXX: ???
+		// we can skip this poke if resource hasn't done work yet... it
+		// needs to be poked if already running, or not running though!
+		// TODO: does this need an || activity flag?
+		if n.Res.GetState() != resources.ResStateProcess {
 			if g.Flags.Debug {
 				log.Printf("%s[%s]: Poke: %s[%s]", v.Kind(), v.GetName(), n.Kind(), n.GetName())
 			}
@@ -80,9 +79,7 @@ func (g *Graph) Poke(v *Vertex, activity bool) error {
 				defer wg.Done()
 				//edge := g.Adjacency[v][nn] // lookup
 				//notify := edge.Notify && edge.Refresh()
-				nn.SendEvent(event.EventPoke, nil)
-				// TODO: check return value?
-				return nil // never error for now...
+				return nn.SendEvent(event.EventPoke, nil)
 			}(n)
 
 		} else {
@@ -91,6 +88,7 @@ func (g *Graph) Poke(v *Vertex, activity bool) error {
 			}
 		}
 	}
+	// TODO: do something with return values?
 	wg.Wait() // wait for all the pokes to complete
 	return nil
 }
@@ -100,13 +98,13 @@ func (g *Graph) BackPoke(v *Vertex) {
 	// these are all the vertices pointing TO v, eg: ??? -> v
 	for _, n := range g.IncomingGraphVertices(v) {
 		x, y, s := v.GetTimestamp(), n.GetTimestamp(), n.Res.GetState()
-		// if the parent timestamp needs poking AND it's not in state
-		// ResStateEvent, then poke it. If the parent is in ResStateEvent it
+		// If the parent timestamp needs poking AND it's not running
+		// Process, then poke it. If the parent is in ResStateProcess it
 		// means that an event is pending, so we'll be expecting a poke
 		// back soon, so we can safely discard the extra parent poke...
 		// TODO: implement a stateLT (less than) to tell if something
 		// happens earlier in the state cycle and that doesn't wrap nil
-		if x >= y && (s != resources.ResStateEvent && s != resources.ResStateCheckApply) {
+		if x >= y && (s != resources.ResStateProcess && s != resources.ResStateCheckApply) {
 			if g.Flags.Debug {
 				log.Printf("%s[%s]: BackPoke: %s[%s]", v.Kind(), v.GetName(), n.Kind(), n.GetName())
 			}
@@ -158,7 +156,8 @@ func (g *Graph) Process(v *Vertex) error {
 	if g.Flags.Debug {
 		log.Printf("%s[%s]: Process()", obj.Kind(), obj.GetName())
 	}
-	obj.SetState(resources.ResStateEvent)
+	defer obj.SetState(resources.ResStateNil) // reset state when finished
+	obj.SetState(resources.ResStateProcess)
 	var ok = true
 	var applied = false // did we run an apply?
 	// is it okay to run dependency wise right now?
@@ -174,8 +173,6 @@ func (g *Graph) Process(v *Vertex) error {
 		log.Printf("%s[%s]: OKTimestamp(%v)", obj.Kind(), obj.GetName(), v.GetTimestamp())
 	}
 
-	obj.SetState(resources.ResStateCheckApply)
-
 	// connect any senders to receivers and detect if values changed
 	if updated, err := obj.SendRecv(obj); err != nil {
 		return errwrap.Wrapf(err, "could not SendRecv in Process")
@@ -201,6 +198,9 @@ func (g *Graph) Process(v *Vertex) error {
 	refresh = g.RefreshPending(v) // do i need to perform a refresh?
 	obj.SetRefresh(refresh)       // tell the resource
 
+	// changes can occur after this...
+	obj.SetState(resources.ResStateCheckApply)
+
 	// check cached state, to skip CheckApply; can't skip if refreshing
 	if !refresh && obj.IsStateOK() {
 		checkOK, err = true, nil
@@ -283,7 +283,7 @@ func (g *Graph) Process(v *Vertex) error {
 		// nodes might fail due to having a too old timestamp!
 		v.UpdateTimestamp()                    // this was touched...
 		obj.SetState(resources.ResStatePoking) // can't cancel parent poke
-		if err := g.Poke(v, activity); err != nil {
+		if err := g.Poke(v); err != nil {
 			return errwrap.Wrapf(err, "the Poke() failed")
 		}
 	}
@@ -343,6 +343,12 @@ func (g *Graph) Worker(v *Vertex) error {
 					break Loop // no event, so no ack!
 				}
 
+				// if process started, but no action yet, skip!
+				if v.Res.GetState() == resources.ResStateProcess {
+					ev.ACK() // ready for next message
+					continue
+				}
+
 				// if running, we skip running a new execution!
 				// if waiting, we skip running a new execution!
 				if running || waiting {

resources/exec.go

@@ -163,7 +163,6 @@ func (obj *ExecRes) Watch(processChan chan *event.Event) error {
 	}
 
 	for {
-		obj.SetState(ResStateWatching) // reset
 		select {
 		case text := <-bufioch:
 			cuid.SetConverged(false)

resources/file.go

@@ -170,7 +170,6 @@ func (obj *FileRes) Watch(processChan chan *event.Event) error {
 			log.Printf("%s[%s]: Watching: %s", obj.Kind(), obj.GetName(), obj.Path) // attempting to watch...
 		}
 
-		obj.SetState(ResStateWatching) // reset
 		select {
 		case event, ok := <-obj.recWatcher.Events():
 			if !ok { // channel shutdown

resources/hostname.go

@@ -138,7 +138,6 @@ func (obj *HostnameRes) Watch(processChan chan *event.Event) error {
 	var send = false // send event?
 
 	for {
-		obj.SetState(ResStateWatching) // reset
 		select {
 		case <-signals:
 			cuid.SetConverged(false)

resources/msg.go

@@ -150,7 +150,6 @@ func (obj *MsgRes) Watch(processChan chan *event.Event) error {
 	var send = false // send event?
 	var exit *error
 	for {
-		obj.SetState(ResStateWatching) // reset
 		select {
 		case event := <-obj.Events():
 			cuid.SetConverged(false)

resources/noop.go

@@ -74,7 +74,6 @@ func (obj *NoopRes) Watch(processChan chan *event.Event) error {
 	var send = false // send event?
 	var exit *error
 	for {
-		obj.SetState(ResStateWatching) // reset
 		select {
 		case event := <-obj.Events():
 			cuid.SetConverged(false)

resources/nspawn.go

@@ -140,7 +140,6 @@ func (obj *NspawnRes) Watch(processChan chan *event.Event) error {
 	var exit *error
 
 	for {
-		obj.SetState(ResStateWatching)
 		select {
 		case event := <-buschan:
 			// process org.freedesktop.machine1 events for this resource's name

resources/password.go

@@ -191,7 +191,6 @@ func (obj *PasswordRes) Watch(processChan chan *event.Event) error {
 	var send = false // send event?
 	var exit *error
 	for {
-		obj.SetState(ResStateWatching) // reset
 		select {
 		// NOTE: this part is very similar to the file resource code
 		case event, ok := <-obj.recWatcher.Events():

resources/pkg.go

@@ -142,7 +142,6 @@ func (obj *PkgRes) Watch(processChan chan *event.Event) error {
 			log.Printf("%s: Watching...", obj.fmtNames(obj.getNames()))
 		}
 
-		obj.SetState(ResStateWatching) // reset
 		select {
 		case event := <-ch:
 			cuid.SetConverged(false)

resources/resources.go

@@ -45,10 +45,9 @@ type ResState int
 // Each ResState should be set properly in the relevant part of the resource.
 const (
 	ResStateNil        ResState = iota
-	ResStateWatching
-	ResStateEvent // an event has happened, but we haven't poked yet
-	ResStateCheckApply
-	ResStatePoking
+	ResStateProcess             // we're in process, but we haven't done much yet
+	ResStateCheckApply          // we're about to run CheckApply
+	ResStatePoking              // we're done CheckApply, and we're about to poke
 )
 
 const refreshPathToken = "refresh"
@@ -535,7 +534,6 @@ func (obj *BaseRes) Poll(processChan chan *event.Event) error {
 	var send = false
 	var exit *error
 	for {
-		obj.SetState(ResStateWatching)
 		select {
 		case <-ticker.C: // received the timer event
 			log.Printf("%s[%s]: polling...", obj.Kind(), obj.GetName())

resources/svc.go

@@ -153,7 +153,6 @@ func (obj *SvcRes) Watch(processChan chan *event.Event) error {
 				set.Remove(svc) // no return value should ever occur
 			}
 
-			obj.SetState(ResStateWatching) // reset
 			select {
 			case <-buschan: // XXX: wait for new units event to unstick
 				cuid.SetConverged(false)
@@ -177,7 +176,6 @@ func (obj *SvcRes) Watch(processChan chan *event.Event) error {
 			}
 
 			log.Printf("Watching: %s", svc) // attempting to watch...
-			obj.SetState(ResStateWatching)  // reset
 			select {
 			case event := <-subChannel:
 

resources/timer.go

@@ -92,7 +92,6 @@ func (obj *TimerRes) Watch(processChan chan *event.Event) error {
 	var send = false
 
 	for {
-		obj.SetState(ResStateWatching)
 		select {
 		case <-obj.ticker.C: // received the timer event
 			send = true

test/shell/t5b.sh

@@ -0,0 +1,7 @@
+#!/bin/bash -e
+
+# should take slightly more than 35s, but fail if we take 45s)
+timeout --kill-after=45s 40s ./mgmt run --yaml t5.yaml --converged-timeout=5 --no-watch --tmp-prefix &
+pid=$!
+wait $pid	# get exit status
+exit $?

test/shell/t5b.yaml

@@ -0,0 +1,84 @@
+---
+graph: mygraph
+comment: simpler exec fan in to fan out example to demonstrate optimization
+resources:
+  exec:
+  - name: exec1
+    cmd: sleep 10s
+    shell: ''
+    timeout: 0
+    watchcmd: ''
+    watchshell: ''
+    ifcmd: ''
+    ifshell: ''
+    pollint: 0
+    state: present
+  - name: exec2
+    cmd: sleep 10s
+    shell: ''
+    timeout: 0
+    watchcmd: ''
+    watchshell: ''
+    ifcmd: ''
+    ifshell: ''
+    pollint: 0
+    state: present
+  - name: exec3
+    cmd: sleep 10s
+    shell: ''
+    timeout: 0
+    watchcmd: ''
+    watchshell: ''
+    ifcmd: ''
+    ifshell: ''
+    pollint: 0
+    state: present
+  - name: exec4
+    cmd: sleep 10s
+    shell: ''
+    timeout: 0
+    watchcmd: ''
+    watchshell: ''
+    ifcmd: ''
+    ifshell: ''
+    pollint: 0
+    state: present
+  - name: exec5
+    cmd: sleep 10s
+    shell: ''
+    timeout: 0
+    watchcmd: ''
+    watchshell: ''
+    ifcmd: ''
+    ifshell: ''
+    pollint: 0
+    state: present
+edges:
+- name: e1
+  from:
+    kind: exec
+    name: exec1
+  to:
+    kind: exec
+    name: exec3
+- name: e2
+  from:
+    kind: exec
+    name: exec2
+  to:
+    kind: exec
+    name: exec3
+- name: e3
+  from:
+    kind: exec
+    name: exec3
+  to:
+    kind: exec
+    name: exec4
+- name: e4
+  from:
+    kind: exec
+    name: exec3
+  to:
+    kind: exec
+    name: exec5