Mailing-List: contact user-help@cassandra.apache.org; run by ezmlm
Precedence: bulk
Reply-To: user@cassandra.apache.org
Received-SPF: pass (nike.apache.org: domain of sylvain@datastax.com designates
 74.125.83.44 as permitted sender)
MIME-Version: 1.0
In-Reply-To: <4DCC04F0.3060707@panasiangroup.com>
References: <4DCBBFAA.5040608@panasiangroup.com>
	<4DCC04F0.3060707@panasiangroup.com>
Date: Thu, 12 May 2011 18:51:44 +0200
Message-ID: <BANLkTimCjP0HZi9_AbxfahRMfaZXZTw3ew@mail.gmail.com>
Subject: Re: Cassandra causing very high load on CPU's 0.6.3
From: Sylvain Lebresne <sylvain@datastax.com>
To: user@cassandra.apache.org
Content-Type: text/plain; charset=ISO-8859-1

On Thu, May 12, 2011 at 6:04 PM, Ali Ahsan <ali.ahsan@panasiangroup.com> wrote:
> On 05/12/2011 04:08 PM, Ali Ahsan wrote:
>>
>> Hi All
>>
>> I am experience some problem with me two Cassandra node with RF=2,Both
>> node CPU's usage is very high,load average: 9.47, 5.72, 5.11 and this
>> causing my application to time out .I have xeon with 8 processor and 16 GB
>> of Ram.and LVM setup for Cassandra.How can i trace the main issue of load i
>> have no swap and swappies also set to 0.I am running Centos 5.5 64 Bit.
>
> Add to this i am using openjdk not sunjdk will this be an issue ?

It is indeed advised to use sunjdk as openjdk is a bit behind as far
as bug fixes are
concerned.

Moreover, 0.6.3 is pretty old now and we do have fixed a number of
issue related to
load spikes, so before investigating further the best advice I can
give you is to upgrade
(either to 0.6.13 if you really feel like staying on 0.6, or to 0.7.5).

--
Sylvain