[
https://issues.apache.org/jira/browse/YARN-1879?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13971980#comment-13971980
]
Xuan Gong commented on YARN-1879:
---------------------------------
[~ozawa] Thanks for taking this.
I have one general question: what do we expect for using RetryCache ? Do we use
the RetryCache to handle the potential duplication operations in RM failover/RM
restart scenarios ?
If this is the case, it might not work as we expected. When failover happens or
RM restarts, the ApplicationMaster service will be re-launched. Based on the
patch you provided, we will re-initiate the RetryCache. Then How can we decide
whether this operation has been processed previously or not ? If we really want
to use RetryCache method, we should figure out a way to rebuild RetryCache,
also need to figure out what information we need to save for RetryCache.
Please correct me if I miss something.
> 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-1879.2-wip.patch, YARN-1879.2.patch, YARN-1879.3.patch,
> YARN-1879.4.patch, YARN-1879.5.patch
>
>
--
This message was sent by Atlassian JIRA
(v6.2#6252)
