[
https://issues.apache.org/jira/browse/FLINK-21602?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17297445#comment-17297445
]
Till Rohrmann commented on FLINK-21602:
---------------------------------------
I think the bigger problem is recovering the checkpoints from remote storage.
This can be potentially take quite some time since it involves I/O.
> Creation of ExecutionGraph happens in main thread
> -------------------------------------------------
>
> Key: FLINK-21602
> URL: https://issues.apache.org/jira/browse/FLINK-21602
> Project: Flink
> Issue Type: Sub-task
> Components: Runtime / Coordination
> Affects Versions: 1.13.0
> Reporter: Till Rohrmann
> Assignee: Till Rohrmann
> Priority: Critical
> Fix For: 1.13.0
>
>
> Currently, the {{AdaptiveScheduler}} creates the {{ExecutionGraph}} in the
> main thread. This also means that we are recovering state in the
> {{JobMasters's}} main thread. This can lead to instabilities because we are
> blocking the main thread for too long. I think the creation of the
> {{ExecutionGraph}} should happen in an {{ioExecutor}}.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
