[
https://issues.apache.org/jira/browse/YARN-8352?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Zhizhen Hou updated YARN-8352:
------------------------------
Description: I submit a job to the yarn, and both two times the AM is
allocated on the same node. After the first allocate call to scheduler, the
follows call should include the black node list, but now the black node list is
always null. (was: I submit a job to the yarn, and both two times the AM is
allocated on the same node. After the first allocate call to scheduler, the
follows call should include the black node list, but now the black node list is
a constant null.)
> AM should retry on a different node after the previous application attempt
> fail
> -------------------------------------------------------------------------------
>
> Key: YARN-8352
> URL: https://issues.apache.org/jira/browse/YARN-8352
> Project: Hadoop YARN
> Issue Type: Improvement
> Components: resourcemanager
> Affects Versions: 2.7.5
> Reporter: Zhizhen Hou
> Priority: Major
>
> I submit a job to the yarn, and both two times the AM is allocated on the
> same node. After the first allocate call to scheduler, the follows call
> should include the black node list, but now the black node list is always
> null.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org
