hadoop-yarn-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Varun Saxena (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (YARN-4644) TestRMRestart fails on YARN-2928 branch
Date Wed, 27 Jan 2016 07:11:39 GMT

    [ https://issues.apache.org/jira/browse/YARN-4644?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15118777#comment-15118777
] 

Varun Saxena commented on YARN-4644:
------------------------------------

As per my analysis. findbugs issue is due to an unused variable(as the report indicates).
This is because the signature of RMNodeStatusEvent constructor has changed after rebase from
trunk. Now instead of passing keep alive app ids' in the constructor, we merely pass the {{NodeStatus}}
object in the constructor from which keep alive app ids' are eventually fetched.

So merely removing the declaration of keepAliveApps should be enough.

I can either fix it here or in YARN-4645. Thoughts ?
I can probably run findbugs in all the impacted projects and check if findbugs is coming anywhere
else due to rebase. Unlikely because otherwise it would show up in report for YARN-4238

> TestRMRestart fails on YARN-2928 branch
> ---------------------------------------
>
>                 Key: YARN-4644
>                 URL: https://issues.apache.org/jira/browse/YARN-4644
>             Project: Hadoop YARN
>          Issue Type: Sub-task
>          Components: timelineserver
>    Affects Versions: YARN-2928
>            Reporter: Varun Saxena
>            Assignee: Varun Saxena
>         Attachments: YARN-4644-YARN-2928.01.patch
>
>
> This was reported by YARN-4238 QA report. Refer to https://builds.apache.org/job/PreCommit-YARN-Build/10389/testReport/
> Error reported is as under :
> {noformat}
> org.mockito.exceptions.verification.TooManyActualInvocations: 
> noOpSystemMetricPublisher.appCreated(
>     <any>,
>     <any>
> );
> Wanted 3 times:
> -> at org.apache.hadoop.yarn.server.resourcemanager.TestRMRestart.testRMRestartGetApplicationList(TestRMRestart.java:955)
> But was 6 times. Undesired invocation:
> -> at org.apache.hadoop.yarn.server.resourcemanager.ResourceManager.recover(ResourceManager.java:1274)
> 	at org.apache.hadoop.yarn.server.resourcemanager.TestRMRestart.testRMRestartGetApplicationList(TestRMRestart.java:955)
> {noformat}
> Failing because in {{RMAppImpl#recover}}, {{sendATSCreateEvent}} has been called twice.

> Has been introduced during rebase I guess.
> After removing the duplicate call, the test passes.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Mime
View raw message