hadoop-yarn-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Junping Du (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (YARN-1506) Replace set resource change on RMNode/SchedulerNode directly with event notification.
Date Thu, 17 Apr 2014 20:59:16 GMT

    [ https://issues.apache.org/jira/browse/YARN-1506?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13973400#comment-13973400
] 

Junping Du commented on YARN-1506:
----------------------------------

Thanks [~bikassaha] for comments! Below is my reply:
bq. Does a node update forced by the admin need to be persisted across RM restarts/failover?
Currently no, it could be addressed later in YARN-998 for RM restart or NM re-registering.

bq. UpdateNodeResourceWhenUnusableTransition should also update the resource IMO so that its
got the correct value when the node becomes usable again. Also, It may be useful to take the
common pieces of code between update when running/not-running to be in a shared place. E.g.
the update of the node resource member var.
Ok. It could be some cases that updating unhealthy node's resource. If so, agree we can abstract
some common code piece there.

bq. Do we need to remember in the RMNode that the node resource was forced to a certain value.
E.g. when a bad node re-registers with the RM then it may send its real resource which could
override the admin value.
Also will be addressed in YARN-998, not only for bad node re-register case, but also for intended
NM restart.

bq. Since the capacity scheduler is the default scheduler that runs in every test, it would
be good to have a test for it.
Yes. Will repeat the same story now in FifoScheduler to CapacityScheduler. Thx!

> Replace set resource change on RMNode/SchedulerNode directly with event notification.
> -------------------------------------------------------------------------------------
>
>                 Key: YARN-1506
>                 URL: https://issues.apache.org/jira/browse/YARN-1506
>             Project: Hadoop YARN
>          Issue Type: Sub-task
>          Components: nodemanager, scheduler
>            Reporter: Junping Du
>            Assignee: Junping Du
>         Attachments: YARN-1506-v1.patch, YARN-1506-v2.patch, YARN-1506-v3.patch, YARN-1506-v4.patch,
YARN-1506-v5.patch, YARN-1506-v6.patch, YARN-1506-v7.patch, YARN-1506-v8.patch, YARN-1506-v9.patch
>
>
> According to Vinod's comments on YARN-312 (https://issues.apache.org/jira/browse/YARN-312?focusedCommentId=13846087&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-13846087),
we should replace RMNode.setResourceOption() with some resource change event.



--
This message was sent by Atlassian JIRA
(v6.2#6252)

Mime
View raw message