You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@hbase.apache.org by "huaxiang sun (JIRA)" <ji...@apache.org> on 2016/10/24 19:10:58 UTC

[jira] [Commented] (HBASE-16936) TestRateLimiter#testOverconsumptionFixedIntervalRefillStrategy is flaky

    [ https://issues.apache.org/jira/browse/HBASE-16936?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15602914#comment-15602914 ] 

huaxiang sun commented on HBASE-16936:
--------------------------------------

Hi [~dimaspivak], mind I take this one? I had another Throttle flakey test to take a look and this is related, thanks.

> TestRateLimiter#testOverconsumptionFixedIntervalRefillStrategy is flaky
> -----------------------------------------------------------------------
>
>                 Key: HBASE-16936
>                 URL: https://issues.apache.org/jira/browse/HBASE-16936
>             Project: HBase
>          Issue Type: Bug
>          Components: test
>            Reporter: Dima Spivak
>
> Seeing this once every month or two in-house. Looks like it's a timing-based test, which makes it prone to flakiness, but I've noticed that whenever it fails, it fails with the same {{AssertionError}} (including values), so it'd be worth digging into. In our case:
> {noformat}
> expected:<1000> but was:<999>
> Stack Trace:
> java.lang.AssertionError: expected:<1000> but was:<999>
>         at org.apache.hadoop.hbase.quotas.TestRateLimiter.testOverconsumptionFixedIntervalRefillStrategy(TestRateLimiter.java:119)
> {noformat}



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