hadoop-yarn-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Siddharth Seth (JIRA)" <j...@apache.org>
Subject [jira] [Comment Edited] (YARN-71) Ensure/confirm that the NodeManager cleans up local-dirs on restart
Date Fri, 22 Mar 2013 02:09:17 GMT

    [ https://issues.apache.org/jira/browse/YARN-71?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13609803#comment-13609803
] 

Siddharth Seth edited comment on YARN-71 at 3/22/13 2:07 AM:
-------------------------------------------------------------

Couple more comments. Hopefully the last set.
- "//queue deletions here, rather than NM init?" - this comment isn't required anymore.
- The log messages in the delete method should not suppress the exception (
LOG.warn("Failed to delete localDir: " + localDir, e) ; )
- There's a log message in the test which says "named nobody" - this likely needs to be user
- Bunch of System.out, and printStackTrace() which need to be removed
- ContainerManager.getContainerStatus does not imply much - ant state except non-final states
will show up as RUNNING. What should be verified is the internal state of the Container -
via container.getContainerState(). You'll need access to the nmcontext in the test to access
this

When testing, did you try different users with the LCE ?
                
      was (Author: sseth):
    Couple more comments. Hopefully the last set.
- "//queue deletions here, rather than NM init?" - this comment isn't required anymore.
- The log messages in the delete method should not suppress the exception (
LOG.warn("Failed to delete localDir: " + localDir, e);)
- There's a log message in the test which says "named nobody" - this likely needs to be user
- Bunch of System.out, and printStackTrace() which need to be removed
- ContainerManager.getContainerStatus does not imply much - ant state except non-final states
will show up as RUNNING. What should be verified is the internal state of the Container -
via container.getContainerState(). You'll need access to the nmcontext in the test to access
this

When testing, did you try different users with the LCE ?
                  
> Ensure/confirm that the NodeManager cleans up local-dirs on restart
> -------------------------------------------------------------------
>
>                 Key: YARN-71
>                 URL: https://issues.apache.org/jira/browse/YARN-71
>             Project: Hadoop YARN
>          Issue Type: Bug
>          Components: nodemanager
>            Reporter: Vinod Kumar Vavilapalli
>            Assignee: Xuan Gong
>            Priority: Critical
>         Attachments: YARN-71.10.patch, YARN-71.11.patch, YARN-71.12.patch, YARN-71.1.patch,
YARN-71.2.patch, YARN-71.3.patch, YARN.71.4.patch, YARN-71.5.patch, YARN-71.6.patch, YARN-71.7.patch,
YARN-71.8.patch, YARN-71.9.patch
>
>
> We have to make sure that NodeManagers cleanup their local files on restart.
> It may already be working like that in which case we should have tests validating this.

--
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