[
https://issues.apache.org/jira/browse/FLINK-35483?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17850577#comment-17850577
]
Junrui Li commented on FLINK-35483:
-----------------------------------
The issue comes from cases in BatchJobRecoveryTest that require waiting for a
batch job to finish recovering. Right now, the only way to move forward is to
check if the executionGraph leaves the 'Reconciling' state after recovery
starts. But our current setup can't accurately detect when the recovery starts,
leading to errors. I'll prepare a PR to fix this.
> BatchJobRecoveryTest related to JM failover produced no output for 900 second
> -----------------------------------------------------------------------------
>
> Key: FLINK-35483
> URL: https://issues.apache.org/jira/browse/FLINK-35483
> Project: Flink
> Issue Type: Bug
> Components: Build System / CI
> Affects Versions: 1.20.0
> Reporter: Weijie Guo
> Assignee: Junrui Li
> Priority: Major
>
> testRecoverFromJMFailover
> https://dev.azure.com/apache-flink/apache-flink/_build/results?buildId=59919&view=logs&j=0da23115-68bb-5dcd-192c-bd4c8adebde1&t=24c3384f-1bcb-57b3-224f-51bf973bbee8&l=9476
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
