You are viewing a plain text version of this content. The canonical link for it is here.
Posted to yarn-issues@hadoop.apache.org by "Weiwei Yang (JIRA)" <ji...@apache.org> on 2018/08/27 03:10:00 UTC
[jira] [Comment Edited] (YARN-8555) Parameterize
TestSchedulingRequestContainerAllocation(Async) to cover both PC handler
options
[ https://issues.apache.org/jira/browse/YARN-8555?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16593149#comment-16593149 ]
Weiwei Yang edited comment on YARN-8555 at 8/27/18 3:09 AM:
------------------------------------------------------------
Looks like processor handler still has some limitation to support some of PC forms, such as multi allocation tags, see more in the validation code in {{MemoryPlacementConstraintManager}}. This is causing some of UT could not pass under the \{{processor}} path.
So \{{scheduler}} path gives a more complete support for various of PCs, \{{processor}} path is more limited right now. Will need to take some more time getting them consistent.
was (Author: cheersyang):
Looks like processor handler still has some limitation to support some of PC forms, such as multi allocation tags, see more in the validation code in \{{MemoryPlacementConstraintManager}}.
> Parameterize TestSchedulingRequestContainerAllocation(Async) to cover both PC handler options
> ----------------------------------------------------------------------------------------------
>
> Key: YARN-8555
> URL: https://issues.apache.org/jira/browse/YARN-8555
> Project: Hadoop YARN
> Issue Type: Sub-task
> Reporter: Weiwei Yang
> Priority: Minor
> Labels: newbie
>
> Current test cases in this 2 classes are only targeting for 1 handler type, \{{scheduler}} or \{{processor}}. Once YARN-8015 is done, we should modify them to be parameterized in order to cover both cases.
--
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