[
https://issues.apache.org/jira/browse/FLINK-26114?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17509649#comment-17509649
]
Matthias Pohl commented on FLINK-26114:
---------------------------------------
I created two new issues that influences the solution for FLINK-26114. We
should adapt the shutdown method for the {{CheckpointIDCounter}} and
{{CompletedCheckpointStore}} to return a {{CompletableFuture}}.
This can be then processed in the scheduler implementations or passed forward
through closeAsync call of the scheduler implementations.
> DefaultScheduler fails fatally in case of an error when shutting down the
> checkpoint-related resources
> ------------------------------------------------------------------------------------------------------
>
> Key: FLINK-26114
> URL: https://issues.apache.org/jira/browse/FLINK-26114
> Project: Flink
> Issue Type: Bug
> Components: Runtime / Coordination
> Affects Versions: 1.15.0, 1.16.0
> Reporter: Matthias Pohl
> Assignee: Niklas Semmler
> Priority: Major
>
> In contrast to the {{AdaptiveScheduler}}, the {{DefaultScheduler}} fails
> fatally in case of an error while cleaning up the checkpoint-related
> resources. This contradicts our new approach of retrying the cleanup of
> job-related data (see FLINK-25433). Instead, we would want the
> {{DefaultScheduler}} to return an exceptionally completed future with the
> exception. This enables the {{DefaultResourceCleaner}} to trigger a retry.
--
This message was sent by Atlassian Jira
(v8.20.1#820001)
