[
https://issues.apache.org/jira/browse/MAPREDUCE-4387?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13404197#comment-13404197
]
Kihwal Lee commented on MAPREDUCE-4387:
---------------------------------------
The test was calling {{ReourceManager#stop()}} after it thought it's done. This
hit {{GenericEventHandler}} with an interrupt while it was trying to enqueue an
event. This bubbled up and hit the RM's {{EventProcessor}} loop, which did
System.exit(-1). It checks whether the JVM is being shutdown, but this is
before {{ShutdownHookManager}} is activated.
It seems {{EventProcessor}} shouldn't do exit(-1) if it's got an exception
during shutdown.
> RM gets fatal error and exits during TestRM
> -------------------------------------------
>
> Key: MAPREDUCE-4387
> URL: https://issues.apache.org/jira/browse/MAPREDUCE-4387
> Project: Hadoop Map/Reduce
> Issue Type: Bug
> Components: resourcemanager
> Affects Versions: 2.0.0-alpha
> Reporter: Kihwal Lee
> Fix For: 2.0.1-alpha, 3.0.0
>
>
> It doesn't happen on my desktop, but it happens frequently during the builds
> with clover enabled. Surefire will report it as fork failure.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
