[
https://issues.apache.org/jira/browse/HBASE-21647?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16742669#comment-16742669
]
Jingyun Tian commented on HBASE-21647:
--------------------------------------
[~stack] getChildrenLatchWithoutLocking will be called when SplitWALProcedure
and TransitRegionStateProcedure done. Thus I think it will have some influence
on the performance when number of regions or WALs is large. Maybe I should do a
performance test to verify this?
The reason I introduce updateState is that when the SCP is doing WAL splitting
or regions assignment, it invokes many child procedures and set the state to
its next state. Thus if I call getCurrentState() to get the state, when doing
WAL splitting, UI will show state as SERVER_CRASH_DELETE_SPLIT_WALS_DIR instead
of SERVER_CRASH_SPLIT_LOGS. That's why when child procedures done, the
updateState flag is false, only when the SCP get scheduled again, the state
could be updated.
> Add status track for splitting WAL tasks
> ----------------------------------------
>
> Key: HBASE-21647
> URL: https://issues.apache.org/jira/browse/HBASE-21647
> Project: HBase
> Issue Type: Sub-task
> Reporter: Jingyun Tian
> Assignee: Jingyun Tian
> Priority: Major
> Attachments: HBASE-21647.master.001.patch, Screenshot from 2019-01-03
> 15-31-26.png, Screenshot from 2019-01-03 15-31-36.png
>
>
> add status track to help operator check the status of splitting WAL
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
