[
https://issues.apache.org/jira/browse/MAPREDUCE-5900?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14005331#comment-14005331
]
Mayank Bansal commented on MAPREDUCE-5900:
------------------------------------------
Thanks [~jlowe] For pointing this out.
Yes this is a duplicate issue of MAPREDUCE-5848
I already commented
https://issues.apache.org/jira/browse/MAPREDUCE-5848?focusedCommentId=14005269&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-14005269
I think we don't need TA_PREEMPTED event for this diagnostics info as we get
that as a part of RM diagnostics.
I have tested this with this small patch and it works fine with new event.
One more issue is that we are targetting this jira for 2.4.1 and MAPREDUCE-5848
depends on MAPREDUCE-5196 which is a very big patch for 2.4.1
I think we should still commit this JIRA in 2.4.1 and should target 5848( if
needed with TA_PREEMPTED even) with 5196 in major release.
Thoughts?
Thanks,
Mayank
> Container preemption interpreted as task failures and eventually job failures
> ------------------------------------------------------------------------------
>
> Key: MAPREDUCE-5900
> URL: https://issues.apache.org/jira/browse/MAPREDUCE-5900
> Project: Hadoop Map/Reduce
> Issue Type: Sub-task
> Components: applicationmaster, mr-am, mrv2
> Affects Versions: 2.4.1
> Reporter: Mayank Bansal
> Assignee: Mayank Bansal
> Attachments: MAPREDUCE-5900-1.patch
>
>
> We have Added preemption exit code needs to be incorporated
> MR needs to recognize the special exit code value of -102 and interpret it as
> a container being killed instead of a container failure.
--
This message was sent by Atlassian JIRA
(v6.2#6252)
