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.
This commit is contained in:
27
test/shell/etcd-conflicting-server.sh
Executable file
27
test/shell/etcd-conflicting-server.sh
Executable file
@@ -0,0 +1,27 @@
|
||||
#!/bin/bash
|
||||
|
||||
. "$(dirname "$0")/../util.sh"
|
||||
|
||||
# run empty graphs, we're just testing etcd clustering
|
||||
$TIMEOUT "$MGMT" run --no-pgp --tmp-prefix empty &
|
||||
pid1=$!
|
||||
sleep 15s # let it startup
|
||||
|
||||
# run a second one that should conflict because a server is already running...
|
||||
$TIMEOUT "$MGMT" run --no-pgp --tmp-prefix empty &
|
||||
pid2=$!
|
||||
wait $pid2
|
||||
e=$?
|
||||
if [ $e -eq 0 ]; then
|
||||
echo "second mgmt exited successfully when error was expected"
|
||||
exit 1
|
||||
fi
|
||||
if [ $e -ne 1 ]; then
|
||||
echo "second mgmt exited with unexpected error of $e"
|
||||
exit $e
|
||||
fi
|
||||
|
||||
$(kill -SIGINT $pid1)& # send ^C to exit 1st mgmt
|
||||
wait $pid1 # get exit status
|
||||
# if pid1 exits because of a timeout, then it blocked, and this is a bug!
|
||||
exit $?
|
||||
Reference in New Issue
Block a user