hbase-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Enis Soztutar (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HBASE-5991) Introduce sequential ZNode based read/write locks
Date Tue, 04 Dec 2012 23:58:59 GMT

    [ https://issues.apache.org/jira/browse/HBASE-5991?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13510139#comment-13510139
] 

Enis Soztutar commented on HBASE-5991:
--------------------------------------

Backporting this (HBASE-5991) and HBASE-5494 to trunk. 
The rational for the read-write lock is that the master-initiate operations (enable/disable,
modify, add/remove column) and region splits should not happen in parallel. However, multiple
region splits can happen concurrently, so using a read/write lock for the table makes sense.
All the table operations initiated from the master take a write lock. And the region splits
take a read lock. 

In this patch, we will only be doing the master write lock parts. I'll open another issue
to track the read locks for region splits. 

One discussion that happened offline in todays HBase hackaton was to use the curator library
from netflix, which already has (supposedly) correct implementations for many zookeeper recipes
(https://github.com/Netflix/curator). The 89-fb patch already contains a read/write zk lock
that is implemented from the zk recipe in pseudo-code, so we have to chose one way or the
other. 

I can go ahead and rebase this patch with the Curator's version to see whether we can go with
it. If we want to continue the discussion on whether to go with curator or not, we can alternatively
create a separate jira. 
                
> Introduce sequential ZNode based read/write locks 
> --------------------------------------------------
>
>                 Key: HBASE-5991
>                 URL: https://issues.apache.org/jira/browse/HBASE-5991
>             Project: HBase
>          Issue Type: Improvement
>            Reporter: Alex Feinberg
>            Assignee: Alex Feinberg
>             Fix For: 0.89-fb
>
>
> This is a continuation of HBASE-5494:
> Currently table-level write locks have been implemented using non-sequential ZNodes as
part of HBASE-5494 and committed to 89-fb branch. This issue is to track converting the table-level
locks to sequential ZNodes and supporting read-write locks, as to solve the issue of preventing
schema changes during region splits or merges.

--
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

Mime
View raw message