hbase-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Ted Yu (Commented) (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HBASE-4508) Backport HBASE-3777 to 0.90 branch
Date Wed, 26 Oct 2011 04:34:32 GMT

    [ https://issues.apache.org/jira/browse/HBASE-4508?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13135694#comment-13135694
] 

Ted Yu commented on HBASE-4508:
-------------------------------

>From Shrijeet:

1. Rolling restart with 0.90 trunk + patch jar : PASSED
2. Verify that 0.90 trunk jar leaks connection with  HTable pool size set to 10 , hbase.zookeeper.property.maxClientCnxns
set to 20 and launching 10 client threads : PASSED
3. Verify that 0.90 trunk jar + patch does not leak connection with  HTable pool size set
to 10 , hbase.zookeeper.property.maxClientCnxns set to 20 and launching 10 client threads
: PASSED
4. Verify that 0.90 trunk jar + patch still holds up when HTable pool size set to 10 , hbase.zookeeper.property.
maxClientCnxns set to 5 and launching 100 client threads : PASSED

The established connections in my final test was 2. This I verified using both netstat and
zookeeper dump in UI.

One caveat: remember to set hbase.connection.per.config false in order to use connection sharing.

5. hbase.connection.per.config=true , max connections to 300 and bumping ulimit from 1024
to 2048, client threads 100 : PASSED
                
> Backport HBASE-3777 to 0.90 branch
> ----------------------------------
>
>                 Key: HBASE-4508
>                 URL: https://issues.apache.org/jira/browse/HBASE-4508
>             Project: HBase
>          Issue Type: Bug
>            Reporter: Ted Yu
>            Assignee: Bright Fulton
>             Fix For: 0.90.5
>
>         Attachments: HBASE-4508.v1.patch, HBASE-4508.v2.patch, HBASE-4508.v3.patch, HBASE-4508.v4.git.patch,
HBASE-4508.v4.patch, HBASE-4508.v5.patch
>
>
> See discussion here: http://search-hadoop.com/m/MJBId1aazTR1/backporting+HBASE-3777+to+0.90&subj=backporting+HBASE+3777+to+0+90
> Rocketfuel has been running 0.90.3 with HBASE-3777 since its resolution.
> They have 10 RS nodes , 1 Master and 1 Zookeeper
> Live writes and reads but super heavy on reads. Cache hit is pretty high.
> The qps on one of their data centers is 50K.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira

        

Mime
View raw message