[
https://issues.apache.org/jira/browse/HBASE-20533?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16492463#comment-16492463
]
Zheng Hu commented on HBASE-20533:
----------------------------------
Got it, see RegionTransitionProcedure#execute(), at #1, we add the region in
to regionInTransition map, and at #2, the remote call failed, we'll put the
region into regionFailedOpen map, then retries 10 times, after the retry
finished, we'll remote the region from regionInTransition map. if we check
the RIT_COUNT_NAME when retrying, then the RIT_COUNT will be 2.
{code}
if (!am.addRegionInTransition(regionNode, this)) { // #1
....
}
case REGION_TRANSITION_DISPATCH:
// 2. send the request to the target server
if (!updateTransition(env, regionNode)) { // #2
// ....
}
{code}
> Fix the flaky TestAssignmentManagerMetrics
> ------------------------------------------
>
> Key: HBASE-20533
> URL: https://issues.apache.org/jira/browse/HBASE-20533
> Project: HBase
> Issue Type: Bug
> Reporter: Zheng Hu
> Assignee: Zheng Hu
> Priority: Major
> Attachments:
> org.apache.hadoop.hbase.master.TestAssignmentManagerMetrics-output.txt,
> org.apache.hadoop.hbase.master.TestAssignmentManagerMetrics.txt
>
>
> See:
> 1.
> https://builds.apache.org/job/HBASE-Find-Flaky-Tests/lastSuccessfulBuild/artifact/dashboard.html
> 2. https://builds.apache.org/job/HBASE-Flaky-Tests/30837/
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
