[
https://issues.apache.org/jira/browse/YARN-5823?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15637682#comment-15637682
]
Arun Suresh commented on YARN-5823:
-----------------------------------
bq. At the current patch, I am doing the guaranteed allocation first, since it
is non-blocking
The reason I had raised YARN-5688 was to ensure the rpc handler for the AM's
allocate is not blocked by the Opportunistic allocator. So, It really does not
matter if the super.allocate() call happens before or after the opportunistic
allocation, since the Scheduler is actually triggered by the node heartbeat
thread.
If you do not see any other issue with the opportunistic allocation happening
before {{super.allocate()}}, I'd prefer we keep things as they are.
> Update NMTokens in case of requests with only opportunistic containers
> ----------------------------------------------------------------------
>
> Key: YARN-5823
> URL: https://issues.apache.org/jira/browse/YARN-5823
> Project: Hadoop YARN
> Issue Type: Bug
> Reporter: Konstantinos Karanasos
> Assignee: Konstantinos Karanasos
> Priority: Blocker
> Attachments: YARN-5823.001.patch, YARN-5823.002.patch
>
>
> At the moment, when an {{AllocateRequest}} contains only opportunistic
> {{ResourceRequests}}, the updated NMTokens are not properly added to the
> {{AllocateResponse}}.
> In such a case the AM does not get back the needed NMTokens that are required
> to start the opportunistic containers at the respective nodes.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org
