[
https://issues.apache.org/jira/browse/HBASE-7848?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13609618#comment-13609618
]
Jonathan Hsieh commented on HBASE-7848:
---------------------------------------
nit: should probably return the value of super.prepare() in case the base
class changes in the future somehow.
{code}
+ public EnabledTableSnapshotHandler prepare() throws Exception {
+ super.prepare();
+ this.tableLock.acquire(); //after this, you should ensure to release this
lock in
+ //case of exceptions
+ return this;
+ }
{code}
I'm thinking about why we only do this on the enabled table snapshot but not on
the disabled table snapshot. I think we would want to block the enables (or
enables and then alters) from happening while a disabled table snapshot is
happening.
If we add to the disabled case, I think we are missing a call to prepare()
here, and might want to move the prepare() implementation to
TakeSnapshotHandler.java.
https://github.com/apache/hbase/blob/trunk/hbase-server/src/main/java/org/apache/hadoop/hbase/master/snapshot/SnapshotManager.java#L431
Thoughts?
> Use ZK-based read/write lock to make flush-type snapshot robust against table
> enable/disable/alter
> --------------------------------------------------------------------------------------------------
>
> Key: HBASE-7848
> URL: https://issues.apache.org/jira/browse/HBASE-7848
> Project: HBase
> Issue Type: Improvement
> Components: snapshots
> Affects Versions: 0.96.0
> Reporter: Ted Yu
> Assignee: Ted Yu
> Attachments: 7848-v1.txt, 7848-v2.txt, hbase-7848_v2.patch
>
>
> Current region split following flush would fail snapshot.
> We can utilize ZK-based read/write lock to make flush-type snapshot robust
--
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
