From owner-freebsd-current@FreeBSD.ORG Wed Dec 22 11:43:12 2004 Return-Path: Delivered-To: freebsd-current@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 2777216A4CE for ; Wed, 22 Dec 2004 11:43:12 +0000 (GMT) Received: from obh.snafu.de (obh.snafu.de [213.73.92.34]) by mx1.FreeBSD.org (Postfix) with ESMTP id 4B88D43D45 for ; Wed, 22 Dec 2004 11:43:11 +0000 (GMT) (envelope-from ob@gruft.de) Received: from ob by obh.snafu.de with local (Exim 4.43 (FreeBSD)) id 1Ch4tZ-000GUq-Tt for current@freebsd.org; Wed, 22 Dec 2004 12:43:10 +0100 Date: Wed, 22 Dec 2004 12:43:09 +0100 From: Oliver Brandmueller To: current@freebsd.org Message-ID: <20041222114309.GB39637@e-Gitt.NET> References: <41C78F9B.4010804@michaelmeltzer.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <41C78F9B.4010804@michaelmeltzer.com> User-Agent: Mutt/1.5.6i Sender: Oliver Brandmueller Subject: Re: twa driver, 3ware 9500s-4lp, speed issue X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Discussions about the use of FreeBSD-current List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 22 Dec 2004 11:43:12 -0000 Hi. On Mon, Dec 20, 2004 at 09:51:07PM -0500, Michael Meltzer wrote: > I have a 3ware 9500s-4lp controller with 4 10,000rpm raptors hooked up > to it. 0+1 configuration. AMD dual 64 bit processor. We had reproducably very bad experiences with 8500 and 9500 3ware disk controllers in combination with Raptors. After a while (this can be hours or weeks) one of the disks was detached from the RAID ("drive timeout"). You could remove it with tw_cli, when trying to rescan it was not found, until the disk was physically removed or the machine was switched off (only rebooting was not enough): This means the "failed" disk needed to be powercycled. This behaviour has been seen on 6 different machines with either 8500 oder 9500 3ware controllers. The acoustic management of the disks was disabled (this seemed to have caused problems also for other people). The disks were OK, the SMART data showed no damage and it was alsways a random disk failing in the RAIDs. This behaviour seems to only show up in an environment, where you have continuous disk I/O with high stepping rates, such as very loaded mailservers. We switched to the slightly more expensive ICP Vortex controllers in our new machines and will be replacing the old 3wares in the older machines. - The behaviour is _NOT_ seen on servers with less disk I/O. - The behaviour is _NOT_ seen on serevrs with other disks and 3wares. If you ever see a drive timeout in this setup think of this posting. - Oliver -- | Oliver Brandmueller | Offenbacher Str. 1 | Germany D-14197 Berlin | | Fon +49-172-3130856 | Fax +49-172-3145027 | WWW: http://the.addict.de/ | | Ich bin das Internet. Sowahr ich Gott helfe. | | Eine gewerbliche Nutzung aller enthaltenen Adressen ist nicht gestattet! |