From owner-freebsd-stable@FreeBSD.ORG Sat Aug 18 08:09:14 2007 Return-Path: Delivered-To: freebsd-stable@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id D957716A419 for ; Sat, 18 Aug 2007 08:09:14 +0000 (UTC) (envelope-from tom.hurst@clara.net) Received: from spork.qfe3.net (spork.qfe3.net [212.13.207.101]) by mx1.freebsd.org (Postfix) with ESMTP id 9F15B13C442 for ; Sat, 18 Aug 2007 08:09:14 +0000 (UTC) (envelope-from tom.hurst@clara.net) Received: from [81.104.144.87] (helo=voi.aagh.net) by spork.qfe3.net with esmtp (Exim 4.66 (FreeBSD)) (envelope-from ) id 1IMJMu-0003dr-Sz; Sat, 18 Aug 2007 09:09:12 +0100 Received: from freaky by voi.aagh.net with local (Exim 4.67 (FreeBSD)) (envelope-from ) id 1IMJMu-0002Iq-Ox; Sat, 18 Aug 2007 09:09:12 +0100 Date: Sat, 18 Aug 2007 09:09:12 +0100 From: Thomas Hurst To: Vivek Khera Message-ID: <20070818080912.GA6704@voi.aagh.net> Mail-Followup-To: Vivek Khera , FreeBSD Stable References: <31BB09D7-B58A-47AC-8DD1-6BB8141170D8@khera.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: Organization: Not much. User-Agent: Mutt/1.5.16 (2007-06-09) Sender: Thomas Hurst Cc: FreeBSD Stable Subject: Re: large RAID volume partition strategy X-BeenThere: freebsd-stable@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Production branch of FreeBSD source code List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 18 Aug 2007 08:09:14 -0000 * Vivek Khera (vivek@khera.org) wrote: > I'll investigate this option. Does anyone know the stability > reliability of the mpt(4) driver on CURRENT? Is it out of GIANT lock > yet? It was hard to tell from the TODO list if it is entirely free of > GIANT or not. Yes, mpt(4) was made MPSAFE in revision 1.41, about 3 months ago: http://www.freebsd.org/cgi/cvsweb.cgi/src/sys/dev/mpt/mpt.c#rev1.41 I've not seen any stability issues with mpt in either of our test systems, running heavy MySQL load over >20 spindles and a couple of controllers each. > My only fear of this is that once this system is in production, that's > pretty much it. Maintenance windows are about 1 year apart, usually > longer. Best temper your fear with some thorough testing then. If you are going to use ZFS in such a situation, though, I might be strongly tempted to use Solaris instead. Why the long gaps between maintenance? -- Thomas 'Freaky' Hurst http://hur.st/