You are viewing a plain text version of this content. The canonical link for it is here.
Posted to mapreduce-issues@hadoop.apache.org by "Wangda Tan (JIRA)" <ji...@apache.org> on 2014/06/03 12:20:02 UTC
[jira] [Commented] (MAPREDUCE-5900) Container preemption
interpreted as task failures and eventually job failures
[ https://issues.apache.org/jira/browse/MAPREDUCE-5900?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14016365#comment-14016365 ]
Wangda Tan commented on MAPREDUCE-5900:
---------------------------------------
Hi [~mayank_bansal],
Thanks for your patch, I think it almost looks good to me, since the TaskAttemptImpl will not separately treat kill/preempt, both of them will not increase retry number of task.
One comment is, current TaskAttemptImpl already has a PreemptTransition, which may confuse people. In my understanding, this is more like a "proactive preemption" introduced by MAPREDUCE-5916, (See my comment https://issues.apache.org/jira/browse/MAPREDUCE-5196?focusedCommentId=14016357&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-14016357, which is waiting for Carlo's reply). My sugguestion is, either rename this transition to "ProactivePreemptionTransiton" and TA_PREEMPTED to TA_PROACTIVE_PREEMPTED, or we can simply remove this transition and merge logic into KillTransition. Does it make sense to you?
Thanks,
Wangda
> 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, MAPREDUCE-5900-branch-241-2.patch, MAPREDUCE-5900-trunk-1.patch, MAPREDUCE-5900-trunk-2.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)