Received: from malur.postgresql.org ([217.196.149.56]) by arkaria.postgresql.org with esmtp (Exim 4.80) (envelope-from ) id 1YuLrd-0006Dq-I2 for pgsql-admin@arkaria.postgresql.org; Mon, 18 May 2015 14:17:53 +0000 Received: from localhost ([127.0.0.1] helo=postgresql.org) by malur.postgresql.org with smtp (Exim 4.80) (envelope-from ) id 1YuLrd-0004WK-1t for pgsql-admin@arkaria.postgresql.org; Mon, 18 May 2015 14:17:53 +0000 Received: from makus.postgresql.org ([2001:4800:1501:1::229]) by malur.postgresql.org with esmtps (TLS1.2:RSA_AES_256_CBC_SHA256:256) (Exim 4.80) (envelope-from ) id 1YuLrc-0004WE-61 for pgsql-admin@postgresql.org; Mon, 18 May 2015 14:17:52 +0000 Received: from mail4.neteven.com ([5.135.111.119]) by makus.postgresql.org with esmtp (Exim 4.84) (envelope-from ) id 1YuLrU-0007Sk-7y for pgsql-admin@postgresql.org; Mon, 18 May 2015 14:17:50 +0000 Received: from localhost (localhost.localdomain [127.0.0.1]) by mail4.neteven.com (Postfix) with ESMTP id A22641F0169 for ; Mon, 18 May 2015 16:17:42 +0200 (CEST) X-Virus-Scanned: Debian amavisd-new at mail4.neteven.com Received: from mail4.neteven.com ([127.0.0.1]) by localhost (mail4.neteven.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id CyrFNPLadKyP for ; Mon, 18 May 2015 16:17:42 +0200 (CEST) Message-ID: <5559F482.8010401@neteven.com> Date: Mon, 18 May 2015 16:17:38 +0200 From: Thomas SIMON User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:24.0) Gecko/20100101 Thunderbird/24.4.0 MIME-Version: 1.0 To: "pgsql-admin@postgresql.org" Subject: Re: Performances issues with SSD volume ? References: <5559E3AD.7090109@neteven.com> <5559EABB.1060405@neteven.com> In-Reply-To: Content-Type: multipart/alternative; boundary="------------010104050501070908040107" X-Pg-Spam-Score: -1.9 (-) List-Archive: List-Help: List-ID: List-Owner: List-Post: List-Subscribe: List-Unsubscribe: X-Mailing-List: pgsql-admin Precedence: bulk Sender: pgsql-admin-owner@postgresql.org This is a multi-part message in MIME format. --------------010104050501070908040107 Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 8bit Wow, thanks a lot for advice Koray! This is a huge difference indeed ! I will start with upgrading my kernel asap :) Thomas Le 18/05/2015 15:51, Koray Eyidoğan a écrit : > Hi Thomas, > > 3.2 kernel may be #1 cause of your I/O load problem: > > http://www.databasesoup.com/2014/09/why-you-need-to-avoid-linux-kernel-32.html > https://medium.com/postgresql-talk/benchmarking-postgresql-with-different-linux-kernel-versions-on-ubuntu-lts-e61d57b70dd4 > > Have a nice day. > > Koray > > > On Mon, May 18, 2015 at 4:35 PM, Thomas SIMON > wrote: > > Hi matheus, > thanks for reply. > > Kernel is slightly different ; with a 3.10.23-xxxx-std-ipv6-64 > kernel (OVH kernel) on HDD server, and 3.2.65-1+deb7u1 kernel on > SSD server. > I'm using debian wheezy on both servers. > > I didn't changed the 2 parameters you talk about. > > Thanks, > > Thomas > > Le 18/05/2015 15:13, Matheus de Oliveira a écrit : >> >> On Mon, May 18, 2015 at 10:05 AM, Thomas SIMON >> > wrote: >> >> After the change, I had the following behavior, and I don't >> understand why : everything seems to work fine (load is ~6/7, >> when it was previously ~25/30 on the HDD server), so the SSD >> server is faster than the HDD one, and my apps run faster >> too, but after some time (can be 5 minutes or 2 hours), the >> load average increases suddently (can reach 150 !) and does >> not decrease, so postgres and my application are almost >> unusable. (even small requests are in statement timeout) >> >> >> Before we go mad, what is the OS/kernel/distro that you are >> using? Did it change from the previous one? >> >> If Linux, have you disabled transparent_huge_pages and >> zone_reclaim_mode (two known problematic Linux features)? >> >> Best regards, >> -- >> Matheus de Oliveira >> Analista de Banco de Dados >> Dextra Sistemas - MPS.Br nível F! >> www.dextra.com.br/postgres >> > > --------------010104050501070908040107 Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: 8bit
Wow, thanks a lot for advice Koray!
This is a huge difference indeed !
I will start with upgrading my kernel asap :) 
Thomas
Le 18/05/2015 15:51, Koray Eyidoğan a écrit :

On Mon, May 18, 2015 at 4:35 PM, Thomas SIMON <tsimon@neteven.com> wrote:
Hi matheus,
thanks for reply.

Kernel is slightly different ; with a 3.10.23-xxxx-std-ipv6-64 kernel (OVH kernel) on HDD server, and  3.2.65-1+deb7u1 kernel on SSD server.
I'm using debian wheezy on both servers.

I didn't changed the 2 parameters you talk about.

Thanks, 
Thomas
Le 18/05/2015 15:13, Matheus de Oliveira a écrit :

On Mon, May 18, 2015 at 10:05 AM, Thomas SIMON <tsimon@neteven.com> wrote:
After the change, I had the following behavior, and I don't understand why : everything seems to work fine (load is ~6/7, when it was previously ~25/30 on the HDD server), so the SSD server is faster than the HDD one, and my apps run faster too, but after some time (can be 5 minutes or 2 hours), the load average increases suddently (can reach 150 !) and does not decrease, so postgres and my application are almost unusable. (even small requests are in statement timeout)

Before we go mad, what is the OS/kernel/distro that you are using? Did it change from the previous one?

If Linux, have you disabled transparent_huge_pages and zone_reclaim_mode (two known problematic Linux features)?

Best regards,
--
Matheus de Oliveira
Analista de Banco de Dados
Dextra Sistemas - MPS.Br nível F!
www.dextra.com.br/postgres




--------------010104050501070908040107--