[ https://issues.apache.org/jira/browse/FLINK-13497?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16900758#comment-16900758 ]
Biao Liu commented on FLINK-13497: ---------------------------------- To [~pnowojski], I can't agree more. I have another concern, the reason of why {{CheckpointCoordinator}} is multi-threaded currently. Is there ever a performance issue with single threaded? Executing the {{CheckpointCoordinator}} in main thread executor makes things easier. But I'm afraid there is too much burden of main thread executor. Using an independent thread is another alternative. > Checkpoints can complete after CheckpointFailureManager fails job > ----------------------------------------------------------------- > > Key: FLINK-13497 > URL: https://issues.apache.org/jira/browse/FLINK-13497 > Project: Flink > Issue Type: Bug > Components: Runtime / Checkpointing > Affects Versions: 1.9.0, 1.10.0 > Reporter: Till Rohrmann > Priority: Critical > Fix For: 1.9.0 > > > I think that we introduced with FLINK-12364 an inconsistency wrt to job > termination a checkpointing. In FLINK-9900 it was discovered that checkpoints > can complete even after the {{CheckpointFailureManager}} decided to fail a > job. I think the expected behaviour should be that we fail all pending > checkpoints once the {{CheckpointFailureManager}} decides to fail the job. -- This message was sent by Atlassian JIRA (v7.6.14#76016)