hadoop-mapreduce-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Siddharth Seth (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (MAPREDUCE-5094) Disable mem monitoring by default in MiniMRYarnCluster
Date Fri, 22 Mar 2013 21:05:15 GMT

    [ https://issues.apache.org/jira/browse/MAPREDUCE-5094?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13611245#comment-13611245
] 

Siddharth Seth commented on MAPREDUCE-5094:
-------------------------------------------

Killed for exceeding vmem. PMEM is <300MB iirc, vmem going over 6GB - ideally, this should
be investigated further.

For now, proposing changing behaviour for MiniMRCluster memory monitoring back to what it
was in 1.x (my guess is enabling the mem checks in 1.x would cause the same behaviour).
                
> Disable mem monitoring by default in MiniMRYarnCluster
> ------------------------------------------------------
>
>                 Key: MAPREDUCE-5094
>                 URL: https://issues.apache.org/jira/browse/MAPREDUCE-5094
>             Project: Hadoop Map/Reduce
>          Issue Type: Bug
>            Reporter: Siddharth Seth
>            Assignee: Siddharth Seth
>
> YARN-449. Some hbase tests were failing since containers were getting killed. 
> I believe these checks are disabled by default on the branch-1 MiniMRCluster.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira

Mime
View raw message