[
https://issues.apache.org/jira/browse/YARN-2249?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Jian He updated YARN-2249:
--------------------------
Attachment: YARN-2249.1.patch
Instead of making client side changes, changed RM to cache the outstanding
release request. And the container won't be recovered if the container remains
in the cache. The cache will be cleaned after NM expire interval if no such
container is received by RM for recovery.
Uploaded a patch based on that.
> RM may receive container release request on AM resync before container is
> actually recovered
> --------------------------------------------------------------------------------------------
>
> Key: YARN-2249
> URL: https://issues.apache.org/jira/browse/YARN-2249
> Project: Hadoop YARN
> Issue Type: Sub-task
> Components: resourcemanager
> Reporter: Jian He
> Assignee: Jian He
> Attachments: YARN-2249.1.patch
>
>
> AM resync on RM restart will send outstanding container release requests back
> to the new RM. In the meantime, NMs report the container statuses back to RM
> to recover the containers. If RM receives the container release request
> before the container is actually recovered in scheduler, the container won't
> be released and the release request will be lost.
--
This message was sent by Atlassian JIRA
(v6.2#6252)
