hbase-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Jean-Marc Spaggiari (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HBASE-10089) Metrics intern table names cause eventual permgen OOM in 0.94
Date Thu, 12 Dec 2013 14:56:08 GMT

    [ https://issues.apache.org/jira/browse/HBASE-10089?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13846341#comment-13846341

Jean-Marc Spaggiari commented on HBASE-10089:

Ran some tests...

bash script doing create table with 15 regions, disable, drop. 1000 times.

338m22s without the patch.
338m13s with the patch.

Did not faced any OOME.

I think most of those 338 minutes are spent on the shell more than the create table... Should
I try with a client api call with HBaseAdmin to try to do that faster? I can do 10 threads,
create/disable/drop too. And see.

Just ask ;)


> Metrics intern table names cause eventual permgen OOM in 0.94
> -------------------------------------------------------------
>                 Key: HBASE-10089
>                 URL: https://issues.apache.org/jira/browse/HBASE-10089
>             Project: HBase
>          Issue Type: Bug
>    Affects Versions: 0.94.0, 0.94.14
>            Reporter: Dave Latham
>            Assignee: Ted Yu
>            Priority: Minor
>             Fix For: 0.94.15
>         Attachments: 10089-0.94.txt
> As part of the metrics system introduced in HBASE-4768 there are two places that hbase
uses String interning ( SchemaConfigured and SchemaMetrics ).  This includes interning table
names.  We have long running environment where we run regular integration tests on our application
using hbase.  Those tests create and drop tables with new names regularly.  These leads to
filling up the permgen with interned table names.  Workaround is to periodically restart the
region servers.

This message was sent by Atlassian JIRA

View raw message