[
https://issues.apache.org/jira/browse/HBASE-23771?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17026462#comment-17026462
]
Michael Stack commented on HBASE-23771:
---------------------------------------
Pushed attached patch on branch-2. It cycles waiting on clearance of
references, waiting on any outstanding compactions first and then scheduling
new ones until references are gone and we can move on. Leaving open while watch
what happens over next 24 hours. Need better tooling/apis for watching over
compactions. Compaction logs are a little baffling and should report explicitly
for example why they choose not to major compact a store file (Will file an
issue).
> [Flakey Tests] Test TestSplitTransactionOnCluster Again
> -------------------------------------------------------
>
> Key: HBASE-23771
> URL: https://issues.apache.org/jira/browse/HBASE-23771
> Project: HBase
> Issue Type: Sub-task
> Reporter: Michael Stack
> Priority: Major
> Attachments:
> 0001-HBASE-23771-Flakey-Tests-Test-TestSplitTransactionOn.patch
>
>
> Parent fix had the test failures in GCE go from 35% to 4%. Let me see if can
> clear the remaining fails.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
