[
https://issues.apache.org/jira/browse/FLINK-5076?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15701748#comment-15701748
]
ASF GitHub Bot commented on FLINK-5076:
---------------------------------------
Github user tillrohrmann commented on the issue:
https://github.com/apache/flink/pull/2817
Thanks for your contribution @ifndef-SleePy. Changes look good to me. As a
follow up, we should rework the `JM`, `RM` and `TM` shutdown to wait for their
termination via `RpcEndpoint#getTerminationEndpoint`, because `shutDown`
triggers a lazy shutdown. I've create a [follow
up](https://issues.apache.org/jira/browse/FLINK-5176) for this.
Merging this.
> Shutting down TM when shutting down new mini cluster
> ----------------------------------------------------
>
> Key: FLINK-5076
> URL: https://issues.apache.org/jira/browse/FLINK-5076
> Project: Flink
> Issue Type: Bug
> Components: Cluster Management
> Environment: FLIP-6 feature branch
> Reporter: Biao Liu
> Assignee: Biao Liu
> Priority: Minor
>
> Currently we don't shut down task manager when shutting down mini cluster. It
> will cause mini cluster can not exit normally.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
