[
https://issues.apache.org/jira/browse/HBASE-25145?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17206138#comment-17206138
]
Duo Zhang commented on HBASE-25145:
-----------------------------------
For normal replication, only the recovered replication source could drain all
the files in the queue, so if we find out there is no more wals then we will
close the replication source.
I do not think the most important thing here is to write a test and 'fix'. You
should have a plan first. In general, we will not close the meta wal provider
in the current code base, so we will not meet this situation as we will always
have a wal file in queue(always empty, though). If you implement the logic to
close the meta wal provider when meta region is gone, you will meet this, but
since you have done this, why not also remove the replication source and add it
back when meta region is back?
> WALReader quits if nothing to replicate (and won't restart)
> -----------------------------------------------------------
>
> Key: HBASE-25145
> URL: https://issues.apache.org/jira/browse/HBASE-25145
> Project: HBase
> Issue Type: Sub-task
> Reporter: Michael Stack
> Priority: Major
>
> Noticed by [~huaxiangsun] in his review of HBASE-25055 "Add ReplicationSource
> for meta WALs; add enable/disable w…"
> {quote}bq. Eventually, the meta wal file will be gced and there is no more
> logs in the queue. In that case, the walReader thread will quit. When the
> meta region is moved back, it does not seem that walReader thread will be
> restarted. So it seems that something is broken.
> {quote}
> This issue is about writing a test to run the above scenario and fix any
> probs. found.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
