From owner-freebsd-scsi@FreeBSD.ORG Tue Mar 22 02:21:36 2005 Return-Path: Delivered-To: freebsd-scsi@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id AB2A916A4CE for ; Tue, 22 Mar 2005 02:21:36 +0000 (GMT) Received: from gw.catspoiler.org (217-ip-163.nccn.net [209.79.217.163]) by mx1.FreeBSD.org (Postfix) with ESMTP id 394AF43D39 for ; Tue, 22 Mar 2005 02:21:36 +0000 (GMT) (envelope-from truckman@FreeBSD.org) Received: from FreeBSD.org (mousie.catspoiler.org [192.168.101.2]) by gw.catspoiler.org (8.13.1/8.13.1) with ESMTP id j2M2LS0v078118; Mon, 21 Mar 2005 18:21:32 -0800 (PST) (envelope-from truckman@FreeBSD.org) Message-Id: <200503220221.j2M2LS0v078118@gw.catspoiler.org> Date: Mon, 21 Mar 2005 18:21:28 -0800 (PST) From: Don Lewis To: rloef@interfold.com In-Reply-To: <20050321183410.T19351@auden.jmla.com> MIME-Version: 1.0 Content-Type: TEXT/plain; charset=us-ascii cc: freebsd-scsi@FreeBSD.org Subject: Re: Any changes in SCSI code between 4.10 and 4.11? X-BeenThere: freebsd-scsi@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: SCSI subsystem List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 22 Mar 2005 02:21:36 -0000 On 21 Mar, Reed Loefgren wrote: > All, > > I have just finished rolling my desktop back to 4.10(p5) from > 4.11RELEASE. It had been suffering freezes, the last one bad enough > that boot time couldn't find /kernel and corrupted my striped vinum > /usr. > > I had fallen back from 4-STABLE to 4.11RELEASE to try and stop this > because things seemed to go bad around the time I moved to 4.11 from > 4.10, but I don't really track stuff like that. I'm using an Adaptec > 2940U(W?) card; 69-ext, 68-int and 50-int connectors and wondered if > there had been any changes to the SCSI code in the move between 4.10 > and 4.11. I imagine the 2940 might also be rather EOL for the codebase > going forward. In any event, it's nice to have again a machine that > stays up more than a day at a time; it's still a great OS. > > I've included no dumps or debug info here, I just thought I'd ask (and I'm > not sure I want to go back to that just to find out something I can now > avoid.) I've seen something similar with one of the disks on my 4-STABLE box. The working -> non-working timeframe sounds similar. I thought the disk was going bad. It tended to fail under load, but I was able to retrieve all the data from it and I replaced it with a couple of older and smaller disks that I had as spares. The symtoms were timeouts and this disk would sometimes go offline, resulting in an "Invalidating pack" error. It also exhibited the same problems on my 6-CURRENT box until just recently, which makes me think that something in the CAM, SCSI, or driver code recently got fixed. I'm now totally unable to reproduce the problem on my 6-CURRENT box and I'm planning on moving the disk back to by 4-STABLE box to see if the problem comes back. In my case, the controller is a ahc0: port 0xe000-0xe0ff mem 0xef022000-0xef022fff irq 16 at device 12.0 on pci0 and the disk is a da1 at ahc0 bus 0 target 1 lun 0 da1: Fixed Direct Access SCSI-3 device da1: 160.000MB/s transfers (80.000MHz, offset 31, 16bit), Tagged Queueing Enabled da1: 35242MB (72176566 512 byte sectors: 255H 63S/T 4492C)