[
https://issues.apache.org/jira/browse/HBASE-7977?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13613992#comment-13613992
]
Jean-Marc Spaggiari commented on HBASE-7977:
--------------------------------------------
[~enis], you're right. It's clear and fine for me know. My -1 is no more
relevant.
Just one question. If the cluster is brought down while one operation with a
lock is in progress, what will be the behaviour when the cluster will be
restarted?
Correct me:
- If it was a read lock, splits will still be possible, but master operations
will not be possible anymore until this lock is manually removed.
- If it was a write lock, no more mater operations nor regions operations will
be possible.
Right?
Regarding [[email protected]]'s proposal, I like it! Will it be possible to
differentiate read from write locks?
What will be the options for the operators to clean that? They can only go to
ZK and clear the node?
> Online merge should acquire table lock
> --------------------------------------
>
> Key: HBASE-7977
> URL: https://issues.apache.org/jira/browse/HBASE-7977
> Project: HBase
> Issue Type: Improvement
> Components: master, Region Assignment, regionserver
> Affects Versions: 0.95.0, 0.98.0
> Reporter: Enis Soztutar
> Assignee: Enis Soztutar
> Fix For: 0.95.0, 0.98.0
>
> Attachments: hbase-7977_v1.patch
>
>
> Once online merge (HBASE-7403) is in, we should ensure that we acquire a
> table write lock during the merge.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
