[
https://issues.apache.org/jira/browse/HAMA-936?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Edward J. Yoon updated HAMA-936:
--------------------------------
Description:
Reported summary: Job has finished successfully but no containers. So finally,
job throws FAILED status with timeout exception.
What happen if the getContainerStatuses only returns the current container
statuses? In other words, if the getContainerStatuses doesn't care about the
completed containers, logic of JobImpl.startJob() implementation is very
unstable.
was:What happen if the getContainerStatuses only returns the current
container statuses? In other words, if the getContainerStatuses doesn't care
about the completed containers, logic of JobImpl.startJob() implementation is
very unstable.
> Occasional yarn job fails with timeout exception
> ------------------------------------------------
>
> Key: HAMA-936
> URL: https://issues.apache.org/jira/browse/HAMA-936
> Project: Hama
> Issue Type: Bug
> Components: yarn
> Affects Versions: 0.6.4
> Reporter: Edward J. Yoon
> Assignee: Edward J. Yoon
> Fix For: 0.7.0
>
>
> Reported summary: Job has finished successfully but no containers. So
> finally, job throws FAILED status with timeout exception.
> What happen if the getContainerStatuses only returns the current container
> statuses? In other words, if the getContainerStatuses doesn't care about the
> completed containers, logic of JobImpl.startJob() implementation is very
> unstable.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
