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 "Nemon Lou (JIRA)" <ji...@apache.org> on 2013/10/12 11:32:41 UTC

[jira] [Updated] (MAPREDUCE-5581) killing jobs which have failed causes log missing

     [ https://issues.apache.org/jira/browse/MAPREDUCE-5581?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Nemon Lou updated MAPREDUCE-5581:
---------------------------------

    Description: 
In hive code,when a job failed,they invoke the RunningJob.killJob() API immediately.
>From mapreduce client side,when job is at failed state,the YARNRunner will invoke resMgrDelegate.killApplication to kill that job.And this prevent AM from writing logs to job history server.

  was:
In hive code,when a job failed,they invoke the RunningJob.killJob() API .
>From mapreduce client side,when job is at failed state,the YARNRunner will invoke resMgrDelegate.killApplication to kill that job.And this causes logs missing from job history server.


> killing jobs which have failed causes log missing
> -------------------------------------------------
>
>                 Key: MAPREDUCE-5581
>                 URL: https://issues.apache.org/jira/browse/MAPREDUCE-5581
>             Project: Hadoop Map/Reduce
>          Issue Type: Bug
>          Components: client
>    Affects Versions: 2.1.1-beta
>            Reporter: Nemon Lou
>
> In hive code,when a job failed,they invoke the RunningJob.killJob() API immediately.
> From mapreduce client side,when job is at failed state,the YARNRunner will invoke resMgrDelegate.killApplication to kill that job.And this prevent AM from writing logs to job history server.



--
This message was sent by Atlassian JIRA
(v6.1#6144)