[
https://issues.apache.org/jira/browse/YARN-11612?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17784005#comment-17784005
]
ASF GitHub Bot commented on YARN-11612:
---------------------------------------
slfan1989 commented on code in PR #6260:
URL: https://github.com/apache/hadoop/pull/6260#discussion_r1386482442
##########
hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-common/src/main/java/org/apache/hadoop/yarn/server/uam/UnmanagedApplicationManager.java:
##########
@@ -431,6 +431,7 @@ private void submitUnmanagedApp(ApplicationId appId) throws
YarnException, IOExc
context.setResource(resource);
context.setAMContainerSpec(amContainer);
if (applicationSubmissionContext != null) {
+
context.setApplicationName(applicationSubmissionContext.getApplicationName());
Review Comment:
This is a good idea, but I'm still a little worried. The initial decision
not to assign application name was to quickly differentiate whether the task
originated from a different sub-cluster. If we were to assign a name, it might
make it more challenging to discern if the application belongs to another
sub-cluster.
> [Federation] Fix the name of unmanaged app.
> -------------------------------------------
>
> Key: YARN-11612
> URL: https://issues.apache.org/jira/browse/YARN-11612
> Project: Hadoop YARN
> Issue Type: Improvement
> Reporter: Chenyu Zheng
> Assignee: Chenyu Zheng
> Priority: Major
> Labels: pull-request-available
>
> When app is running across multiple cluster, the app is running in secondary
> cluster as unmanaged app. For now, the app name is
> "UnmanagedAM-${clusterid}", this name should be real app name which is set by
> user.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
