Extends collection of shutdown Reasons (e.g. ClusterJoinUnsuccessfulReason) #26570
Conversation
| /** | ||
| * Scala API: The shutdown was initiated by a failure to join a seed node. | ||
| */ | ||
| object JoiningSeedNodesUnsuccessfulReason extends Reason |
There was a problem hiding this comment.
That seems extremely specific, should it rather be ClusterJoinUnsuccessful or something like that, or is it important that it was specifically joining to seed nodes that failed?
There was a problem hiding this comment.
I agree, could be slightly less specific
| /** | ||
| * Java API: The shutdown was initiated by a configuration clash within the existing cluster and the joining node | ||
| */ | ||
| def incompatibleConfigurationDetectedReason: Reason = IncompatibleConfigurationDetectedReason |
There was a problem hiding this comment.
Sounds like it makes sense to make it public
akka-ci
added
needs-attention
|
Test FAILed. |
akka-ci
added
validating
|
Test FAILed. |
ignasi35
changed the title
| @@ -501,7 +501,7 @@ private[cluster] class ClusterCoreDaemon(publisher: ActorRef, joinConfigCompatCh | |||
| seedNodes.mkString(", "), | |||
| ShutdownAfterUnsuccessfulJoinSeedNodes) | |||
| joinSeedNodesDeadline = None | |||
| CoordinatedShutdown(context.system).run(CoordinatedShutdown.ClusterDowningReason) | |||
| CoordinatedShutdown(context.system).run(CoordinatedShutdown.ClusterJoinUnsuccessfulReason) | |||
There was a problem hiding this comment.
Currently we have ClusterDowningReason configured to use exit code -1 (abnormal termination). Shouldn't ClusterJoinUnsuccessfulReason also be configured to use -1.
I guess -1 will have the effect that an orchestration platform will start it again, which is probably desired also for unsuccessful join?
There was a problem hiding this comment.
Hmmm, I think the exit status was relevant in ConductR but I don't think it is in Kubernetes. IDK about other orchestrators.
I'll fix this PR but I think we could consider deprecating the feature.
There was a problem hiding this comment.
The exit code is useful. k8 isn't the only thing. A bash script can check exit code for decision of restarting process.
There was a problem hiding this comment.
Hmmm, in the case of IncompatibleConfigurationDetectedReason restarting won't help so I don't think a custom exit code is necessary.
There was a problem hiding this comment.
I agree, IncompatibleConfigurationDetectedReason
akka-ci
added
validating
|
Test FAILed. |
akka-ci
added
validating
|
Test FAILed. |
akka-ci
added
needs-attention
|
Test FAILed. |
akka-ci
added
validating
|
Test PASSed. |
|
The Travis job failed this morning but I think the cause was a failing bintray. Can someone relaunch, please? |
|
PLS BUILD |
akka-ci
added
validating
|
Test PASSed. |
Related to lagom/lagom#1806