hadoop-yarn-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Wangda Tan (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (YARN-8292) Fix the dominant resource preemption cannot happen when some of the resource vector becomes negative
Date Thu, 17 May 2018 21:39:00 GMT

    [ https://issues.apache.org/jira/browse/YARN-8292?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16479772#comment-16479772
] 

Wangda Tan commented on YARN-8292:
----------------------------------

Finally found the issue :).

The issue is caused by improper calculation of computeFixpointAllocation / TempQueuePerPartition.offer:


I added several componentWiseMin/Max to avoid DRF computation issues, should be straightforward
to review.

Attached ver.4 patch, [~eepayne], mind to check the patch again?

> Fix the dominant resource preemption cannot happen when some of the resource vector becomes
negative
> ----------------------------------------------------------------------------------------------------
>
>                 Key: YARN-8292
>                 URL: https://issues.apache.org/jira/browse/YARN-8292
>             Project: Hadoop YARN
>          Issue Type: Bug
>          Components: yarn
>            Reporter: Sumana Sathish
>            Assignee: Wangda Tan
>            Priority: Critical
>         Attachments: YARN-8292.001.patch, YARN-8292.002.patch, YARN-8292.003.patch, YARN-8292.004.patch
>
>
> This is an example of the problem: 
>   
> {code}
>     //   guaranteed,  max,    used,   pending
>     "root(=[30:18:6  30:18:6 12:12:6 1:1:1]);" + //root
>         "-a(=[10:6:2 10:6:2  6:6:3   0:0:0]);" + // a
>         "-b(=[10:6:2 10:6:2  6:6:3   0:0:0]);" + // b
>         "-c(=[10:6:2 10:6:2  0:0:0   1:1:1])"; // c
> {code}
> There're 3 resource types. Total resource of the cluster is 30:18:6
> For both of a/b, there're 3 containers running, each of container is 2:2:1.
> Queue c uses 0 resource, and have 1:1:1 pending resource.
> Under existing logic, preemption cannot happen.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

---------------------------------------------------------------------
To unsubscribe, e-mail: yarn-issues-unsubscribe@hadoop.apache.org
For additional commands, e-mail: yarn-issues-help@hadoop.apache.org


Mime
View raw message