From owner-freebsd-stable@FreeBSD.ORG Mon Feb 11 10:37:42 2013 Return-Path: Delivered-To: freebsd-stable@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) by hub.freebsd.org (Postfix) with ESMTP id 55FEA652 for ; Mon, 11 Feb 2013 10:37:42 +0000 (UTC) (envelope-from prvs=07549c782d=ob@gruft.de) Received: from main.mx.e-gitt.net (service.rules.org [IPv6:2001:1560:2342::2]) by mx1.freebsd.org (Postfix) with ESMTP id EFDFD19CA for ; Mon, 11 Feb 2013 10:37:41 +0000 (UTC) Received: from ob by main.mx.e-gitt.net with local (Exim 4.80.1 (FreeBSD)) (envelope-from ) id 1U4qlW-00020e-7S for freebsd-stable@freebsd.org; Mon, 11 Feb 2013 11:37:38 +0100 Date: Mon, 11 Feb 2013 11:37:38 +0100 From: Oliver Brandmueller To: freebsd-stable@freebsd.org Subject: Re: patch which implements ZFS LZ4 compression Message-ID: <20130211103737.GS38901@e-Gitt.NET> Mail-Followup-To: freebsd-stable@freebsd.org References: <511581C9.5040608@delphij.net> <20130208230830.GA45081@icarus.home.lan> <20130209151918.221176cd@fabiankeil.de> <20130209204436.GA67890@icarus.home.lan> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20130209204436.GA67890@icarus.home.lan> X-Face: "TT~P'b_)-jKU_0^a=usXryz`YTz)z.[FZrI,A~PREI2U}frrZ`>_J&; ^t|^.dR/mqtC,Vb.Y>~u8(|aL)vAv(k">zY"]*m*y|b8S7:WK[/qP5i>HO#Ek; C[X:b|FP0*Ly_4Ni User-Agent: Mutt/1.5.21 (2010-09-15) Sender: Oliver Brandmueller X-BeenThere: freebsd-stable@freebsd.org X-Mailman-Version: 2.1.14 Precedence: list List-Id: Production branch of FreeBSD source code List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 11 Feb 2013 10:37:42 -0000 Hi, On Sat, Feb 09, 2013 at 12:44:36PM -0800, Jeremy Chadwick wrote: > Bottom line: people enable compression on an fs, issue large amounts of > write I/O to that fs (say hundreds of megabytes, or gigabytes), and > start to see the entire system intermittently stalling hard (for > multiple seconds at a time). This affects everything from switching VTs > on physical console to packets going across SSH. The stalls vary in > duration depending on what compression type is used (lzjb vs. gzip-1 -- > I cannot even imagine what gzip-9 would be like). I described it as > verbosely as I could, including going back and "re-testing" because > people felt the "ZFSv28 import might have addressed it" (it did not): > > http://lists.freebsd.org/pipermail/freebsd-fs/2011-October/012752.html > > The exact same behaviour happens if dedup is used. There is no relation > between compression (the feature) and dedup (the feature), obviously, > but the symptom I've described matches Bob's explanation perfectly. > > If you want to provide the aforementioned instructions, I'll happily > follow them. Did you try using 4BSD instead of ULE at some point? I had similar problems and that completely fixed it for me. Which would mean, that there's some interactio between scheduler and ZFS code. - Oliver -- | Oliver Brandmueller http://sysadm.in/ ob@sysadm.in | | Ich bin das Internet. Sowahr ich Gott helfe. |