[
https://issues.apache.org/jira/browse/HADOOP-9528?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Steve Loughran resolved HADOOP-9528.
------------------------------------
Resolution: Cannot Reproduce
Fix Version/s: 2.8.0
TestZKFailoverController is a woefully unreliable test, but this particular
test case isn't the one that's causing the trouble these days.
closing as cannot reproduce (for now)
> org.apache.hadoop.ha.TestZKFailoverController.testCedeActive is failing
> intermittently
> ---------------------------------------------------------------------------------------
>
> Key: HADOOP-9528
> URL: https://issues.apache.org/jira/browse/HADOOP-9528
> Project: Hadoop Common
> Issue Type: Bug
> Components: ha
> Affects Versions: 2.0.4-alpha
> Reporter: Konstantin Boudnik
> Fix For: 2.8.0
>
>
> Unit tests is failing with a message like
> {{Should take ~3 seconds to rejoin. Only took 2276ms before rejoining.}}
> A looking into the test shows the only 2.8 seconds and more is considered to
> be {{about 3 seconds}}. Say 2.7 seconds won't be considered {{about 3
> seconds}} contrary to elementary school arithmetic.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
