You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@flink.apache.org by "Robert Metzger (JIRA)" <ji...@apache.org> on 2015/04/10 14:40:12 UTC
[jira] [Commented] (FLINK-1842) ApplicationMaster for YARN FIFO
tests does not shut down properly
[ https://issues.apache.org/jira/browse/FLINK-1842?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14489536#comment-14489536 ]
Robert Metzger commented on FLINK-1842:
---------------------------------------
I just tried to reproduce the issue on my machine but {{ps aux | grep java}} didn't show my any lingering ApplicationMasters.
Did the tests properly pass?
Can you reproduce the issue?
> ApplicationMaster for YARN FIFO tests does not shut down properly
> -----------------------------------------------------------------
>
> Key: FLINK-1842
> URL: https://issues.apache.org/jira/browse/FLINK-1842
> Project: Flink
> Issue Type: Bug
> Components: YARN Client
> Affects Versions: 0.9
> Reporter: Till Rohrmann
>
> I noticed that I had 12 dead java processes consuming 2 GB memory on my machine which weren't shown using {{jps}}. It turned out that all of them were {{ApplicationMaster}} processes. I suspect that this is the ApplicationMaster belonging to the {{YARNSessionFIFOITCase}} test case and that the ApplicationMaster is not properly shutdown after the tests.
> In order to not pollute the developer's system, it would be good to properly terminate all YARN components.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)