[
https://issues.apache.org/jira/browse/HBASE-8803?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13706313#comment-13706313
]
Nick Dimiduk commented on HBASE-8803:
-------------------------------------
My testing has been pained, in part because I'm learning how to use the cluster
resources available to me. Also because I'm frequently hitting troubles where
one of the ROOT or META regions become stuck and unmovable. Perhaps 1/3 runs of
the rolling-restart.sh script experience this issue. I'm not sure if this is a
known issue or something funny about my cluster.
Excluding the times when that happens, my results aren't particularly
encouraging:
{noformat}
# 9 region servers, 12 regions total:
## 0.94.6.1
real 2m58.097s
user 3m22.527s
sys 0m10.687s
real 3m48.006s
user 3m26.695s
sys 0m11.141s
real 2m27.157s
user 3m15.894s
sys 0m9.895s
real 2m24.395s
user 3m12.016s
sys 0m9.766s
## 0.94.6.1 + 8803, maxthreads 1 (default)
real 7m50.688s
user 3m33.115s
sys 0m11.260s
real 4m13.964s
user 3m24.539s
sys 0m10.290s
real 2m57.423s
user 3m22.539s
sys 0m10.143s
## 0.94.6.1 + 8803, maxthreads 10
real 3m30.100s
user 3m23.651s
sys 0m10.185s
real 3m16.714s
user 3m19.534s
sys 0m10.002s
{noformat}
I'll continue to run it in the background. I expect the performance improvement
will be more apparent with more regions.
> region_mover.rb should move multiple regions at a time
> ------------------------------------------------------
>
> Key: HBASE-8803
> URL: https://issues.apache.org/jira/browse/HBASE-8803
> Project: HBase
> Issue Type: Bug
> Components: Usability
> Affects Versions: 0.98.0, 0.94.8, 0.95.1
> Reporter: Jean-Marc Spaggiari
> Assignee: Jean-Marc Spaggiari
> Fix For: 0.98.0, 0.95.2, 0.94.10
>
> Attachments: HBASE-8803-v0-trunk.patch, HBASE-8803-v1-0.94.patch,
> HBASE-8803-v1-trunk.patch, HBASE-8803-v2-0.94.patch
>
> Original Estimate: 48h
> Remaining Estimate: 48h
>
> When there is many regions in a cluster, rolling_restart can take hours
> because region_mover is moving the regions one by one.
--
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
