[
https://issues.apache.org/jira/browse/HBASE-21095?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16588458#comment-16588458
]
Duo Zhang commented on HBASE-21095:
-----------------------------------
OK I think the problem is that I do not implement the timeot logic correctly...
When restarting, if the procedure is still in WAITING_TIMEOUT state, it will be
put into the timeoutExecutor, and after the timeout, in setTimeoutFailure we
will wake the procedure up. But the problem here is that, after restarting, we
will loss the information of the procedures waiting on the event, so we just
wake up nothing and the procedures will hang there forever...
Let me see how we implement the logic before HBASE-20881...
> TestDisableTableProcedure.testRecoveryAndDoubleExecution sometimes hangs
> ------------------------------------------------------------------------
>
> Key: HBASE-21095
> URL: https://issues.apache.org/jira/browse/HBASE-21095
> Project: HBase
> Issue Type: Bug
> Components: amv2, proc-v2
> Reporter: Duo Zhang
> Priority: Major
>
> It also uses TRSP as sub procedure so probably we should set killIfHasParent
> to true, but the log is a bit interesting, that we just hang there without
> executing any procedures after a restart, but for other tests where we need
> to set killIfHasParent to true, we will keep executing procedures but do not
> make any progress.
> Need to dig more.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
