[
https://issues.apache.org/jira/browse/HBASE-7431?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13539132#comment-13539132
]
Lars Hofhansl commented on HBASE-7431:
--------------------------------------
This test also starts *way* to many region servers.
HBaseTestingUtil.ensureSomeNonStoppedRegionServersAvailable is wrong. For each
stopped/stopping RegionServer is starts a new one. As RSs are started/stopped
there will be many that are just stopping, and this method will create way more
running RSs than requested.
> TestSplitTransactionOnCluster tests still flaky
> -----------------------------------------------
>
> Key: HBASE-7431
> URL: https://issues.apache.org/jira/browse/HBASE-7431
> Project: HBase
> Issue Type: Sub-task
> Reporter: Lars Hofhansl
> Assignee: Lars Hofhansl
> Fix For: 0.96.0, 0.94.4
>
> Attachments: 7431-0.94.txt, 7431-0.94-v2.txt
>
>
> I still see occasional failures:
> # Assuming all regions of a table are only, even when a previous may have
> shut down one of the RSs, in which case it might take some for these regions
> to be online again.
> # Assuming propagation from ZK to RS is instant, even though communication to
> watchers is asynchronous
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
