[
https://issues.apache.org/jira/browse/HBASE-19929?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16353564#comment-16353564
]
Duo Zhang commented on HBASE-19929:
-----------------------------------
It is introduced by HBASE-7329, and now I think it is useless. We removed the
WAL marker for flush in HBASE-7329, but now the flush marker is back... Since
we have set closed flag to true, a flush will fail when it wants to write the
flush marker, no matter whether we are waiting before actually close the writer
as there is a check for closed flag in WAL.append.
So I tend to remove the DrainBarrier. In general I do not think this will
introduce any big problems such as data inconsistency.
Thanks.
> Call RS.stop on a session expired RS may hang
> ---------------------------------------------
>
> Key: HBASE-19929
> URL: https://issues.apache.org/jira/browse/HBASE-19929
> Project: HBase
> Issue Type: Bug
> Reporter: Duo Zhang
> Priority: Major
>
> See the discussion in HBASE-19927. The problem is that, for a normal stop we
> will try to close all the regions and wait until they are all closed. But if
> the RS has already session expired, master will start the failover work which
> will move the WAL directory, and then we will be stuck in writing flush
> marker.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
