[
https://issues.apache.org/jira/browse/FLINK-34007?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17806714#comment-17806714
]
Yang Wang commented on FLINK-34007:
-----------------------------------
Maybe this issue is related with fabric8 K8s client behavior change. In
v5.12.4[1], the {{onStartLeading}} callback will always be executed when
acquired the leadership. But in v6.6.2[2], the {{onStartingLeading}} callback
will be executed only when the holder identity changed. In this case, the
JobManager does not crash while the holder identity keeps same.
[1].
[https://github.com/fabric8io/kubernetes-client/blob/v5.12.4/kubernetes-client/src/main/java/io/fabric8/kubernetes/client/extended/leaderelection/LeaderElector.java#L71]
[2].
[https://github.com/fabric8io/kubernetes-client/blob/v6.6.2/kubernetes-client-api/src/main/java/io/fabric8/kubernetes/client/extended/leaderelection/LeaderElector.java#L232]
> Flink Job stuck in suspend state after losing leadership in HA Mode
> -------------------------------------------------------------------
>
> Key: FLINK-34007
> URL: https://issues.apache.org/jira/browse/FLINK-34007
> Project: Flink
> Issue Type: Bug
> Components: Runtime / Coordination
> Affects Versions: 1.16.3, 1.17.2, 1.18.1, 1.18.2
> Reporter: Zhenqiu Huang
> Priority: Major
> Attachments: Debug.log, job-manager.log
>
>
> The observation is that Job manager goes to suspend state with a failed
> container not able to register itself to resource manager after timeout.
> JM Log, see attached
>
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
