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 "Karthik Kambatla (JIRA)" <ji...@apache.org> on 2015/08/19 19:38:46 UTC

[jira] [Commented] (MAPREDUCE-6457) kill task before the task is truely runing on a slave node

    [ https://issues.apache.org/jira/browse/MAPREDUCE-6457?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14703430#comment-14703430 ] 

Karthik Kambatla commented on MAPREDUCE-6457:
---------------------------------------------

Could you share more details on the kinds of errors you saw. If were working on MR internals, may be a test case to demonstrate the issue?

> kill task before the task is truely runing on a slave node
> ----------------------------------------------------------
>
>                 Key: MAPREDUCE-6457
>                 URL: https://issues.apache.org/jira/browse/MAPREDUCE-6457
>             Project: Hadoop Map/Reduce
>          Issue Type: Bug
>          Components: applicationmaster, mrv2
>    Affects Versions: 2.6.0, 2.7.1
>         Environment: regular environment
>            Reporter: Wei Chen
>   Original Estimate: 24h
>  Remaining Estimate: 24h
>
> I happned to find this bug. When I am doing some development based on Hadoop-2.6.0.  I happend to send to T_KILL signal to one of maptask before this task is register to TaskAttemptListener. Because the attempt is not assigned with a container yet. It may lead to a lot of errors.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)