[
https://issues.apache.org/jira/browse/HADOOP-12706?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15097284#comment-15097284
]
Colin Patrick McCabe commented on HADOOP-12706:
-----------------------------------------------
I just reproduced this locally by running the test in a loop until it happened
to fail. Just like you experienced, the system doesn't seem to be heavily
loaded.
I don't know how deterministic {{PhantomReference}} behavior is "supposed" to
be. It could be that the JVM just doesn't feel like it needs to run the GC in
this small unit test, and so the P.R.s never get cleaned up. I am going to try
adding a call to {{System.gc}} when the {{waitFor}} comes up false, to see if
that fixes things.
> TestLocalFsFCStatistics#testStatisticsThreadLocalDataCleanUp times out
> occasionally
> -----------------------------------------------------------------------------------
>
> Key: HADOOP-12706
> URL: https://issues.apache.org/jira/browse/HADOOP-12706
> Project: Hadoop Common
> Issue Type: Bug
> Components: test
> Reporter: Jason Lowe
> Assignee: Sangjin Lee
>
> TestLocalFsFCStatistics has been failing sometimes, and when it fails it
> appears to be from FCStatisticsBaseTest.testStatisticsThreadLocalDataCleanUp.
> The test is timing out when it fails.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
