From owner-freebsd-scsi@FreeBSD.ORG Fri Jan 14 00:38:32 2011 Return-Path: Delivered-To: freebsd-scsi@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 8FB701065672; Fri, 14 Jan 2011 00:38:32 +0000 (UTC) (envelope-from joachim@tingvold.com) Received: from smtp.domeneshop.no (smtp.domeneshop.no [194.63.248.54]) by mx1.freebsd.org (Postfix) with ESMTP id 4C4C08FC17; Fri, 14 Jan 2011 00:38:32 +0000 (UTC) Received: from aannecy-552-1-175-159.w83-201.abo.wanadoo.fr ([83.201.191.159] helo=keklolwtf.home) by smtp.domeneshop.no with esmtpsa (TLS1.0:RSA_AES_128_CBC_SHA1:16) (Exim 4.72) (envelope-from ) id 1PdXgU-0000MX-Tg; Fri, 14 Jan 2011 01:38:31 +0100 Mime-Version: 1.0 (Apple Message framework v1076) Content-Type: text/plain; charset=us-ascii; format=flowed; delsp=yes From: Joachim Tingvold In-Reply-To: <20110114001758.GA12793@nargothrond.kdm.org> Date: Fri, 14 Jan 2011 01:38:27 +0100 Content-Transfer-Encoding: 7bit Message-Id: References: <4D2DAA45.30602@FreeBSD.org> <41C64262-4300-4187-B5FD-04A5EFB7F87C@tingvold.com> <20110113203750.GA39494@nargothrond.kdm.org> <20110114001758.GA12793@nargothrond.kdm.org> To: Kenneth D. Merry X-Mailer: Apple Mail (2.1076) Cc: freebsd-scsi@freebsd.org, Alexander Motin Subject: Re: mps0-troubles X-BeenThere: freebsd-scsi@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: SCSI subsystem List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 14 Jan 2011 00:38:32 -0000 On Fri, Jan 14, 2011, at 01:17:58AM GMT+01:00, Kenneth D. Merry wrote: >> This won't "conflict" with what you've written here[1]? The 9211-8i >> controller I have, use the SAS2008-chip. > > No, this affects the number of buffers available for scatter/gather > entries. Most commands will just have one physical memory buffer > attached, > since system buffer memory is physically contiguous. > > It doesn't affect the number of commands the driver sends down to the > chip. I see. The strange thing is that I copied around 7-8TB at around 105MB/s (via samba) to 'storage', without ever having this issue. There has been no hardware-change of any kind since this was done. >> Would it be possible that low amounts of free memory (<100MB, the >> system has a total of 4GB) could cause this? (since I'm using ZFS, >> which relies heavily on memory). > > I don't think so. If anything, I think that this is likely > triggered by > a large number of outstanding commands, or perhaps a leak > somewhere. If > it's the former, hopefully this will fix it. If it's the latter, > you'll > eventually run into the problem again. Okay. I've changed the value. I'll keep you posted. -- Joachim