You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@hbase.apache.org by "Jimmy Xiang (JIRA)" <ji...@apache.org> on 2012/12/03 19:43:59 UTC
[jira] [Created] (HBASE-7261) batchMutate/put release the same lock
twice
Jimmy Xiang created HBASE-7261:
----------------------------------
Summary: batchMutate/put release the same lock twice
Key: HBASE-7261
URL: https://issues.apache.org/jira/browse/HBASE-7261
Project: HBase
Issue Type: Bug
Reporter: Jimmy Xiang
Assignee: Jimmy Xiang
Put uses batchMutate to do the actual work. Put calls startRegionOperation/closeRegionOperation. BatchMutate does the same. If the same thread already holds the lock, lock it again doesn't increase the lock count.
However, releasing lock is a little different. If the lock is already released, IllegalMonitorStateException will throw if it is released again.
There could be other calls. I will look into it more.
--
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
[jira] [Commented] (HBASE-7261) batchMutate/put release the same
lock twice
Posted by "Jimmy Xiang (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-7261?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13508959#comment-13508959 ]
Jimmy Xiang commented on HBASE-7261:
------------------------------------
This happens in 0.94 too. In 0.94, put and internalObtainRowLock call startRegionOperation/closeRegionOperation twice in the same thread.
Something surprising me is that the IllegalMonitorStateException doesn't happen all the time. Did I miss anything?
> batchMutate/put release the same lock twice
> -------------------------------------------
>
> Key: HBASE-7261
> URL: https://issues.apache.org/jira/browse/HBASE-7261
> Project: HBase
> Issue Type: Bug
> Reporter: Jimmy Xiang
> Assignee: Jimmy Xiang
>
> Put uses batchMutate to do the actual work. Put calls startRegionOperation/closeRegionOperation. BatchMutate does the same. If the same thread already holds the lock, lock it again doesn't increase the lock count.
> However, releasing lock is a little different. If the lock is already released, IllegalMonitorStateException will throw if it is released again.
> There could be other calls. I will look into it more.
--
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
[jira] [Resolved] (HBASE-7261) batchMutate/put release the same
lock twice
Posted by "Jimmy Xiang (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HBASE-7261?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Jimmy Xiang resolved HBASE-7261.
--------------------------------
Resolution: Invalid
It is an issue with my patch for HBASE-6423 actually. Closed as invalid.
> batchMutate/put release the same lock twice
> -------------------------------------------
>
> Key: HBASE-7261
> URL: https://issues.apache.org/jira/browse/HBASE-7261
> Project: HBase
> Issue Type: Bug
> Reporter: Jimmy Xiang
> Assignee: Jimmy Xiang
>
> Put uses batchMutate to do the actual work. Put calls startRegionOperation/closeRegionOperation. BatchMutate does the same. If the same thread already holds the lock, lock it again doesn't increase the lock count.
> However, releasing lock is a little different. If the lock is already released, IllegalMonitorStateException will throw if it is released again.
> There could be other calls. I will look into it more.
--
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