[
https://issues.apache.org/jira/browse/YARN-422?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13649809#comment-13649809
]
Zhijie Shen commented on YARN-422:
----------------------------------
Like ContainerLauncher, I think NMClient is supposed to start when AM starts
and stop when AM stops. IMHO, no matter how long a container keeps running, it
should be stopped when the AM who starts it stops. In addition, NMClient would
be the only gate to access NodeManager (I mean it's not good to use NMClient
and the raw RPC proxy simultaneously), such that if NMClient doesn't stop the
running containers when NMClient stops (or AM exits), these containers may be
not stoppable. Semantically, it is not good that AM has already stopped, while
the forked containers keep running.
> Add NM client library
> ---------------------
>
> Key: YARN-422
> URL: https://issues.apache.org/jira/browse/YARN-422
> Project: Hadoop YARN
> Issue Type: Sub-task
> Reporter: Bikas Saha
> Assignee: Zhijie Shen
> Attachments: AMNMClient_Defination.txt,
> AMNMClient_Definition_Updated_With_Tests.txt, proposal_v1.pdf,
> YARN-422.1.patch, YARN-422.2.patch, YARN-422.3.patch
>
>
> Create a simple wrapper over the ContainerManager protocol to provide hide
> the details of the protocol implementation.
--
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
