Thick in prod, thin in the replication filer.

From: Jordan Slingerland <Jordan.Slingerland@independenthealth.com>
To: Fred Grieco <fredgrieco@yahoo.com>; Jeff Mohler <speedtoys.racing@gmail.com>
Cc: "Toasters@teaparty.net" <Toasters@teaparty.net>
Sent: Friday, September 13, 2013 3:23 PM
Subject: RE: high CP_reads/writes ratioA

 
At 87% it is certainly possible that wafl doesn’t have enough contagious free space to write a chain in a single pass…slowing things down substantially.
 
Are your volumes thick provisioned?  That could cloud the equation.
 
--JMS
 
 
 
From: Fred Grieco [mailto:fredgrieco@yahoo.com]
Sent: Friday, September 13, 2013 3:16 PM
To: Jordan Slingerland; Jeff Mohler
Cc: Toasters@teaparty.net
Subject: Re: high CP_reads/writes ratio
 
The prod one is 87%.  The replication one is 71% but it was much higher during disaster recovery testing.

I'll run a reallocate measure on the DR side this weekend and see what comes back.
 
 
From: Jordan Slingerland <Jordan.Slingerland@independenthealth.com>
To: Jeff Mohler <speedtoys.racing@gmail.com>; Fred Grieco <fredgrieco@yahoo.com>
Cc: "Toasters@teaparty.net" <Toasters@teaparty.net>
Sent: Friday, September 13, 2013 3:12 PM
Subject: RE: high CP_reads/writes ratio
 
Are your aggregates more then 65-85% full? 
 
From: toasters-bounces@teaparty.net [mailto:toasters-bounces@teaparty.net] On Behalf Of Jeff Mohler
Sent: Friday, September 13, 2013 3:09 PM
To: Fred Grieco
Cc: Toasters@teaparty.net
Subject: Re: high CP_reads/writes ratio
 
Wafl free space fragmentation most likely.  Reallocate all volumes in the affected aggregate.  
 
Do not reallocate the _aggregate_.  

Sent from my iPhone

On Sep 13, 2013, at 1:06 PM, Fred Grieco <fredgrieco@yahoo.com> wrote:
What kinds of things can cause a high CP_reads/writes ratio?  And what can I do about it? 

I have an Oracle system with ASM that doing writes in a 15 minute cycle, and it causes a lot of writes with CP reads.   The oracle ASM luns are basically the only thing on the aggregate.  Disk alignment is fine (which is the first thing I thought of).  I've attached a screenshot of the Oncommand graph for one of the disks in this aggr-- not sure if that will come through.

The filer is a 3220 with 8.1.2 7-mode.  Fibre channel attach, and there are about 20 luns under two volumes.

TIA,

Fred


_______________________________________________
Toasters mailing list
Toasters@teaparty.net
http://www.teaparty.net/mailman/listinfo/toasters