[
https://issues.apache.org/jira/browse/FLINK-16945?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17169680#comment-17169680
]
Dian Fu commented on FLINK-16945:
---------------------------------
An instance from release-1.11:
https://dev.azure.com/apache-flink/apache-flink/_build/results?buildId=5104&view=logs&j=c5f0071e-1851-543e-9a45-9ac140befc32&t=684b1416-4c17-504e-d5ab-97ee44e08a20
KafkaProducerExactlyOnceITCase.testExactlyOnceCustomOperator failed with
"Exceeded checkpoint tolerable failure threshold".
> Execute CheckpointFailureManager.FailJobCallback directly in main thread
> executor
> ---------------------------------------------------------------------------------
>
> Key: FLINK-16945
> URL: https://issues.apache.org/jira/browse/FLINK-16945
> Project: Flink
> Issue Type: Improvement
> Components: Runtime / Checkpointing
> Affects Versions: 1.10.0
> Reporter: Biao Liu
> Assignee: Biao Liu
> Priority: Critical
> Labels: pull-request-available
> Fix For: 1.12.0
>
> Time Spent: 20m
> Remaining Estimate: 0h
>
> Since we have put all non-IO operations of {{CheckpointCoordinator}} into
> main thread executor, the {{CheckpointFailureManager.FailJobCallback}} could
> be executed directly now. In this way execution graph would fail immediately
> when {{CheckpointFailureManager}} invokes the callback. We could avoid the
> inconsistent scenario of FLINK-13497.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
