Received: from makus.postgresql.org (makus.postgresql.org [98.129.198.125])
	by mail.postgresql.org (Postfix) with ESMTP id 22C5E7591C6
	for <pgsql-performance@postgresql.org>;
	Thu, 24 May 2012 09:58:23 -0300 (ADT)
Received: from corp.tradeindia.com ([61.16.154.82]
	helo=zmta01.trade-india-local.com)
	by makus.postgresql.org with esmtp (Exim 4.72)
	(envelope-from <mallah@tradeindia.com>) id 1SXXcU-00028j-5O
	for pgsql-performance@postgresql.org; Thu, 24 May 2012 12:58:23 +0000
Received: from localhost (localhost.localdomain [127.0.0.1])
	by zmta01.trade-india-local.com (Postfix) with ESMTP id A5D6D44E52EE;
	Thu, 24 May 2012 18:28:07 +0530 (IST)
X-Virus-Scanned: amavisd-new at zmta01.trade-india-local.com
Received: from zmta01.trade-india-local.com ([127.0.0.1])
	by localhost (zmta01.trade-india-local.com [127.0.0.1]) (amavisd-new,
	port 10024)
	with ESMTP id 4khmw2i6JUGD; Thu, 24 May 2012 18:28:07 +0530 (IST)
Received: from zmbox01.trade-india-local.com (zmbox01.trade-india-local.com
	[192.168.1.111])
	by zmta01.trade-india-local.com (Postfix) with ESMTP id 6698444E5261;
	Thu, 24 May 2012 18:28:07 +0530 (IST)
Date: Thu, 24 May 2012 18:28:07 +0530 (IST)
From: "Rajesh Kumar. Mallah" <mallah@tradeindia.com>
To: Andy Colson <andy@squeakycode.net>
Cc: Claudio Freire <klaussfreire@gmail.com>, pgsql-performance@postgresql.org
Message-ID: 
 <14870043.444411337864287337.JavaMail.root@zmbox01.trade-india-local.com>
In-Reply-To: 
 <22172945.443861337864106055.JavaMail.root@zmbox01.trade-india-local.com>
Subject: Re: High load average in 64-core server , no I/O wait and CPU is idle
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
X-Originating-IP: [192.168.1.114]
X-Mailer: Zimbra 6.0.1_GA_1816.DEBIAN4.0 (ZimbraWebClient - FF3.0
	(Linux)/6.0.1_GA_1816.DEBIAN4.0)
X-Pg-Spam-Score: 1.4 (+)
X-Archive-Number: 201205/121
X-Sequence-Number: 46945


Dear Andy ,

Following the discussion on load average we are now  investigating on some 
other parts of the stack (other than db). 

Essentially we are bumping up the limits (on appserver) so that more requests 
goes to the DB server.


| 
| Maybe you are hitting some locks?   If its not IO and not CPU then
| maybe something is getting locked and queries are piling up.




| 
| -Andy