[
https://issues.apache.org/jira/browse/FLINK-27053?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17520338#comment-17520338
]
Matthias Pohl commented on FLINK-27053:
---------------------------------------
Hi [~Zsigner], the issue here is that we want to integrate the cleanup of
checkpoints into the retry loop of the Dispatcher job cleanup. That would
require errors to be propagated through exceptions instead of swallowing them
and printing some log message.
> IncrementalRemoteKeyedStateHandle.discardState swallows errors
> --------------------------------------------------------------
>
> Key: FLINK-27053
> URL: https://issues.apache.org/jira/browse/FLINK-27053
> Project: Flink
> Issue Type: Sub-task
> Components: Runtime / Coordination
> Reporter: Matthias Pohl
> Priority: Major
>
> IncrementalRemoteKeyedStateHandle.discardState swallows errors instead of
> propagating them which would make the discard failure go un-noticed.
--
This message was sent by Atlassian Jira
(v8.20.1#820001)
