Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752939AbbEFOCy (ORCPT ); Wed, 6 May 2015 10:02:54 -0400 Received: from cantor2.suse.de ([195.135.220.15]:38683 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752910AbbEFOCw (ORCPT ); Wed, 6 May 2015 10:02:52 -0400 Date: Wed, 6 May 2015 16:02:50 +0200 From: David Sterba To: Toralf =?iso-8859-1?Q?F=F6rster?= Cc: Linux Kernel Subject: Re: periodic iowait spike Message-ID: <20150506140249.GN5099@twin.jikos.cz> Reply-To: dsterba@suse.cz Mail-Followup-To: dsterba@suse.cz, Toralf =?iso-8859-1?Q?F=F6rster?= , Linux Kernel References: <5545F220.1060902@gmx.de> MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Disposition: inline Content-Transfer-Encoding: 8bit In-Reply-To: <5545F220.1060902@gmx.de> User-Agent: Mutt/1.5.23.1-rc1 (2014-03-12) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 896 Lines: 18 On Sun, May 03, 2015 at 12:02:08PM +0200, Toralf F?rster wrote: > The picture in [1], created with isag from the sysstat package, shows > a periodic spike in the iowait of my tinderbox server (hardened Gentoo > 3.19.6-r1), correlating to write transactions per seconds from a > similar graph. > > I'm wondering what (kernel, btrfs or ?) is causing this. Could be btrfs and the transaction commit, 30 seconds by default. The plot roughly matches to the ~30 seconds delay, but not everywhere. Raw numbers would could tell you more. Also, you could try to change the commit interval (mount option 'commit') and see if the iowait spikes change as well. -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/