[
https://issues.apache.org/jira/browse/FLINK-5022?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15636693#comment-15636693
]
ASF GitHub Bot commented on FLINK-5022:
---------------------------------------
GitHub user tillrohrmann opened a pull request:
https://github.com/apache/flink/pull/2757
[FLINK-5022] Suppress RejectedExecutionExceptions if the ExecutorService
has been shut down
This PR suppresses occurring RejectedExecutionExceptions if an
ExecutorService has been shut
down. This only works for ExecutorServices at the moment. All other
exceptions are logged.
You can merge this pull request into a Git repository by running:
$ git pull https://github.com/tillrohrmann/flink
fixRejectedExecutionException
Alternatively you can review and apply these changes as the patch at:
https://github.com/apache/flink/pull/2757.patch
To close this pull request, make a commit to your master/trunk branch
with (at least) the following in the commit message:
This closes #2757
----
commit e400578da826cb5b2d7cb0b4b0518df1b2284248
Author: Till Rohrmann <[email protected]>
Date: 2016-11-04T10:16:47Z
[FLINK-5022] Suppress RejectedExecutionExceptions if the ExecutorService
has been shut down
This PR suppresses occurring RejectedExecutionExceptions if an
ExecutorService has been shut
down. This only works for ExecutorServices at the moment. All other
exceptions are logged.
----
> Suppress RejectedExecutionException when the Executor is shut down
> ------------------------------------------------------------------
>
> Key: FLINK-5022
> URL: https://issues.apache.org/jira/browse/FLINK-5022
> Project: Flink
> Issue Type: Improvement
> Affects Versions: 1.2.0
> Reporter: Till Rohrmann
> Assignee: Till Rohrmann
> Priority: Minor
> Fix For: 1.2.0
>
>
> The {{JobManager}} shuts down its {{ExecutionService}} when the actor is
> stopped. This causes the problem that the execution service which is used to
> run future callbacks is finished before the actor has completely terminated.
> Due to this there might be rpc answers which cause another {{Runnable}} to be
> submitted to the {{ExecutionService}}. This causes
> {{RejectedExecutionExceptions}} to be thrown.
> The underlying problem is that the {{JobManager}} should not stop the
> {{ExecutionService}} before the actor has been terminated. This will be
> changed in Flip-6 anyway. In the meantime to suppress these exceptions, it
> will be checked whether the {{ExecutionService}} has been shut down and only
> if not, exceptions will be logged.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
