From owner-freebsd-stable Fri Mar 7 10:38:39 2003 Delivered-To: freebsd-stable@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id EC7B437B401; Fri, 7 Mar 2003 10:38:37 -0800 (PST) Received: from hub.org (hub.org [64.49.215.141]) by mx1.FreeBSD.org (Postfix) with ESMTP id BC8F443FD7; Fri, 7 Mar 2003 10:38:35 -0800 (PST) (envelope-from scrappy@hub.org) Received: from hub.org (hub.org [64.49.215.141]) by hub.org (Postfix) with ESMTP id AB97E8B6CF2; Fri, 7 Mar 2003 14:38:35 -0400 (AST) Date: Fri, 7 Mar 2003 14:38:35 -0400 (AST) From: "Marc G. Fournier" To: Scott Long Cc: David Schultz , freebsd-stable@freebsd.org Subject: Re: bug in aac driver ... ? In-Reply-To: <3E68B705.4070409@btc.adaptec.com> Message-ID: <20030307143810.O18433@hub.org> References: <20030305224856.R38115@hub.org> <20030307090449.GB61037@HAL9000.homeunix.com> <3E68B705.4070409@btc.adaptec.com> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-stable@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG On Fri, 7 Mar 2003, Scott Long wrote: > David Schultz wrote: > > Thus spake Marc G. Fournier : > > > >> I have an Intel server (SVW7500, or some designation like that) setup > >>with an Adaptec 2120S controller in it: > >> > >>aac0: mem 0xf8000000-0xfbffffff irq 2 at device 9.0 on pci1 > >>aac0: i960RX 100MHz, 48MB cache memory, optional battery present > >>aac0: Kernel 4.0-0, Build 5770, S/N 232fb7 > >> > >> Originally, when we installed it, we had a problem where, while booting, > >>you would suddenly get the 'aac timeout' message scroll up the screen, > >>that has/had since been fixed ... > >> > >> Yesterday, after doing some work on one of our servers to increase the # > >>of vnodes, I added and built a new kernel with: > >> > >>options VM_KMEM_SIZE="(400*1024*1024)" > >>options VM_KMEM_SIZE_MAX="(400*1024*1024)" > >> > >> Which I have running on venus (its using an AMI RAID card) without any > >>hassles ... upon reboot of the Adaptec/Intel server, I once more am > >>getting the Timeouts and inability to boot ... a kernel built 10 minutes > >>earlier, without those options, but the exact same code, runs fine, so > >>something with the VM_KMEM_SIZE increase is causing the problem ... > > > > > > I don't know why VM_KMEM_SIZE would tickle a RAID controller > > problem, but I've seen timeout message similar to the one you > > describe with a flaky SCSI controller. (Try statically compiling > > all relevant drivers into the kernel if you're not already.) The > > fix for me was to reduce the tag depth with camcontrol. Try a > > value around 8 or so; there shouldn't be much, if any, of a > > performance difference. > > > > camcontrol tags devicename -N8 > > > > To Unsubscribe: send mail to majordomo@FreeBSD.org > > with "unsubscribe freebsd-stable" in the body of the message > > The aac driver does not use the CAM SCSI interface for accessing its > RAID arrays, so changing parameters there will have no effect. Oh good, I thought I was doing something wrong when I kept getting errors trying to access it through camcontrol :) To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message