hadoop-yarn-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Thomas Graves (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (YARN-276) Capacity Scheduler can hang when submit many jobs concurrently
Date Wed, 05 Jun 2013 03:55:22 GMT

    [ https://issues.apache.org/jira/browse/YARN-276?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13675574#comment-13675574
] 

Thomas Graves commented on YARN-276:
------------------------------------

I need to spend more time looking through the new logic, here are a few more comments for
now.

remove the comment from overAMUsedPercent about max active application since its no longer
present.

FicaSchedulerApp getAMResource, change amRequedResource -> amRequiredResource

the max active applications per user used to use the absolute queue capacity instead of the
absolute max queue capacity. It was changed to use the absolute capacity because it uses the
userlimitfactor in the calculation, which should be applied to the capacity and not max capacity
(see MAPREDUCE-3897 for more details).  We should change the overAMUsedPercentPerUser similarly
to use absolute capacity, not absolute max capacity.

This can be filed as a separate jira since it was pre-existing but a bad app that requests
0 for the memory could cause divide by zero exception.

                
> Capacity Scheduler can hang when submit many jobs concurrently
> --------------------------------------------------------------
>
>                 Key: YARN-276
>                 URL: https://issues.apache.org/jira/browse/YARN-276
>             Project: Hadoop YARN
>          Issue Type: Bug
>          Components: capacityscheduler
>    Affects Versions: 3.0.0, 2.0.1-alpha
>            Reporter: nemon lou
>            Assignee: nemon lou
>              Labels: incompatible
>         Attachments: YARN-276.patch, YARN-276.patch, YARN-276.patch, YARN-276.patch,
YARN-276.patch, YARN-276.patch, YARN-276.patch, YARN-276.patch, YARN-276.patch, YARN-276.patch,
YARN-276.patch, YARN-276.patch, YARN-276.patch, YARN-276.patch
>
>   Original Estimate: 24h
>  Remaining Estimate: 24h
>
> In hadoop2.0.1,When i submit many jobs concurrently at the same time,Capacity scheduler
can hang with most resources taken up by AM and don't have enough resources for tasks.And
then all applications hang there.
> The cause is that "yarn.scheduler.capacity.maximum-am-resource-percent" not check directly.Instead
,this property only used for maxActiveApplications. And maxActiveApplications is computed
by minimumAllocation (not by Am actually used).

--
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

Mime
View raw message