Mailing-List: contact commits-help@cassandra.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@cassandra.apache.org
Date: Tue, 9 Oct 2012 19:24:02 +0000 (UTC)
From: "John Watson (JIRA)" <jira@apache.org>
To: commits@cassandra.apache.org
Message-ID: <708530416.16253.1349810642877.JavaMail.jiratomcat@arcas>
In-Reply-To: <393176688.90539.1347928868294.JavaMail.jiratomcat@arcas>
Subject: [jira] [Commented] (CASSANDRA-4673) Compaction of HintsColumnFamily
 stucks
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable


    [ https://issues.apache.org/jira/browse/CASSANDRA-4673?page=3Dcom.atlas=
sian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=3D=
13472656#comment-13472656 ]=20

John Watson commented on CASSANDRA-4673:
----------------------------------------

We also experienced this issue on SSDs and after reading CASSANDRA-4492 and=
 CASSANDRA-4182

Turned off {{multithread_compaction}} and this issue went away
               =20
> Compaction of HintsColumnFamily stucks
> --------------------------------------
>
>                 Key: CASSANDRA-4673
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-4673
>             Project: Cassandra
>          Issue Type: Bug
>    Affects Versions: 1.1.1
>         Environment: We've got a 24 nodes cluster with 3 virtual data cen=
ters. We've got 7 nodes with SSD drives. We are operating under very heavy =
read/write load. We typically seeing CPU usage >90% on our machines.
> We are using 1.1.2 - why this version is not listed in the 'Affected Vers=
ion' drop down?
>            Reporter: Bart=C5=82omiej Roma=C5=84ski
>
> On some nodes the compaction of HintsColumnFamily stucked. Here is a typi=
cal output of 'nodetool compactionstats':
> pending tasks: 1
>           compaction type        keyspace   column family bytes compacted=
     bytes total  progress
>                Compaction          systemHintsColumnFamily       34620582=
8       346909662    99.80%
> Active compaction remaining time :   0h00m00s
> Rebooting a node does not help. The compaction starts immediately after b=
ooting and stucks and the same point.
> If this can be related we are also expiring a problem described in CASSAN=
DRA-4639.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrato=
rs
For more information on JIRA, see: http://www.atlassian.com/software/jira