[
https://issues.apache.org/jira/browse/YARN-1879?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14159831#comment-14159831
]
Jian He commented on YARN-1879:
-------------------------------
bq. How about finishApplicationMaster?
ApplicationMasterService#unregisterAttempt is invoked in
RMAppAttempt#BaseFinalTransition. And the BaseFinalTransition will be invoked
only after the am container process exists and NM reports the finished AM
container to RM. This means, if AM is still calling finishApplicationMaster,
the BaseFinalTransition will never be invoked and the responseMap will never be
null for that am.
seems allocate also doesn't need retry-cache, as it internally already
implements retry-cache like mechanism to return previous response for duplicate
request.
> Mark Idempotent/AtMostOnce annotations to ApplicationMasterProtocol
> -------------------------------------------------------------------
>
> Key: YARN-1879
> URL: https://issues.apache.org/jira/browse/YARN-1879
> Project: Hadoop YARN
> Issue Type: Sub-task
> Components: resourcemanager
> Reporter: Jian He
> Assignee: Tsuyoshi OZAWA
> Priority: Critical
> Attachments: YARN-1879.1.patch, YARN-1879.1.patch,
> YARN-1879.11.patch, YARN-1879.12.patch, YARN-1879.13.patch,
> YARN-1879.14.patch, YARN-1879.15.patch, YARN-1879.16.patch,
> YARN-1879.17.patch, YARN-1879.18.patch, YARN-1879.19.patch,
> YARN-1879.2-wip.patch, YARN-1879.2.patch, YARN-1879.20.patch,
> YARN-1879.21.patch, YARN-1879.3.patch, YARN-1879.4.patch, YARN-1879.5.patch,
> YARN-1879.6.patch, YARN-1879.7.patch, YARN-1879.8.patch, YARN-1879.9.patch
>
>
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
