[
https://issues.apache.org/jira/browse/YARN-4643?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15118654#comment-15118654
]
Sidharta Seethana commented on YARN-4643:
-----------------------------------------
The unit test failures are unrelated to this patch. The fix is a one line
change that was manually tested against trunk using a distributed shell app
that stays up through multiple NM restarts.
> Container recovery is broken with delegating container runtime
> --------------------------------------------------------------
>
> Key: YARN-4643
> URL: https://issues.apache.org/jira/browse/YARN-4643
> Project: Hadoop YARN
> Issue Type: Sub-task
> Components: yarn
> Affects Versions: 2.8.0
> Reporter: Sidharta Seethana
> Assignee: Sidharta Seethana
> Priority: Critical
> Attachments: YARN-4643.001.patch
>
>
> Delegating container runtime uses the container's launch context to determine
> which runtime to use. However, during container recovery, a container object
> is not passed as input which leads to a {{NullPointerException}} when
> attempting to access the container's launch context.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
