[
https://issues.apache.org/jira/browse/YARN-8649?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16587944#comment-16587944
]
Jason Lowe commented on YARN-8649:
----------------------------------
Thanks for the analysis and patch, [~xiaoheipangzi]!
Is ignoring the null the right thing to do here? This in the middle of trying
to find a path to localize a resource, and if the NM doesn't know about the
resource then it seems inappropriate to go ahead and find a local path to put
the resource and let the localizer go ahead and download it. That will be a
waste of network and disk resources at best or an outright leak of the disk
space at worst if its not cleaned up when the localizer finishes the download
and reports completion on a resource the NM doesn't know about.
> Similar as YARN-4355:NPE while processing localizer heartbeat
> -------------------------------------------------------------
>
> Key: YARN-8649
> URL: https://issues.apache.org/jira/browse/YARN-8649
> Project: Hadoop YARN
> Issue Type: Bug
> Affects Versions: 3.1.1
> Reporter: lujie
> Assignee: lujie
> Priority: Major
> Attachments: YARN-8649.patch, hadoop-hires-nodemanager-hadoop11.log
>
>
> I have noticed that a nodemanager was getting NPEs while tearing down. The
> reason maybe similar to YARN-4355 which is reported by [# Jason Lowe].
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
