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-7521) fix HBASE-6060 (regions stuck in opening state) in 0.94
Date Thu, 31 Jan 2013 06:35:14 GMT

    [ https://issues.apache.org/jira/browse/HBASE-7521?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13567403#comment-13567403
] 

Enis Soztutar commented on HBASE-7521:
--------------------------------------

I have tested 0.94-tip with and without the patch while running IntegrationTestDataIngestWithChaosMonkey,
which is CM + LoadTestTool. The cluster was a 3 node EC2 running Hadoop1.0 from HDP. Without
the patch, test succeeds in a 5 min run, however with the patch, my 2 attempts failed, with
lots of ServerNotRunningYetException, plus Failed to write key: 3282. Also noticed that it
took way too long to run the ingest (7500 inserts took ~15 min vs ~1:30 with the patch). 

Overall, the patch looks good from my limited understanding of the AM+SSH code base. Everytime
I try to study it, I am running away horrified :) 

One question about the patch is:
why don't we do the same thing in HRegionServer.closeRegion() and set the znode in the RPC
handler?

So, if we want to cut 0.94.5 tomorrow, I guess we can't include this without further testing.

                
> fix HBASE-6060 (regions stuck in opening state) in 0.94
> -------------------------------------------------------
>
>                 Key: HBASE-7521
>                 URL: https://issues.apache.org/jira/browse/HBASE-7521
>             Project: HBase
>          Issue Type: Bug
>            Reporter: Sergey Shelukhin
>            Assignee: Sergey Shelukhin
>             Fix For: 0.94.5
>
>         Attachments: HBASE-7521-original-patch-ported-v0.patch, HBASE-7521-original-patch-ported-v1.patch,
HBASE-7521-original-patch-ported-v2.patch, HBASE-7521-v0.patch, HBASE-7521-v1.patch
>
>
> Discussion in HBASE-6060 implies that the fix there does not work on 0.94. Still, we
may want to fix the issue in 0.94 (via some different fix) because the regions stuck in opening
for ridiculous amounts of time is not a good thing to have.

--
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