From owner-freebsd-stable@FreeBSD.ORG Wed Aug 6 05:42:00 2003 Return-Path: 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 BDAD637B401 for ; Wed, 6 Aug 2003 05:42:00 -0700 (PDT) Received: from pc2-cove3-6-cust88.brhm.cable.ntl.com (pc2-cove3-6-cust88.brhm.cable.ntl.com [81.107.10.88]) by mx1.FreeBSD.org (Postfix) with ESMTP id 35A6243F75 for ; Wed, 6 Aug 2003 05:41:59 -0700 (PDT) (envelope-from ianjhart@ntlworld.com) Received: from gamma.private.lan (gamma.private.lan [192.168.0.12]) id h76CfqxA012106; Wed, 6 Aug 2003 13:41:53 +0100 (BST) (envelope-from ianjhart@ntlworld.com) From: ian j hart To: Scott Long , Sebastien G Date: Wed, 6 Aug 2003 13:41:52 +0100 User-Agent: KMail/1.5.2 References: <3E8F6DEC.50505@alternatives.ca> <3E939680.8050501@alternatives.ca> <3E9398CD.6000303@btc.adaptec.com> In-Reply-To: <3E9398CD.6000303@btc.adaptec.com> MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Content-Disposition: inline Message-Id: <200308061341.52879.ianjhart@ntlworld.com> cc: freebsd-stable@freebsd.org Subject: Re: problems with ultra320 X-BeenThere: freebsd-stable@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Production branch of FreeBSD source code List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 06 Aug 2003 12:42:01 -0000 On Wednesday 09 April 2003 4:51 am, Scott Long wrote: > Sebastien G wrote: > > Well, Thank you Scott. > > > > I set the tag depth to 32 and everything went ok (make buildworld & > > installworld) ! Do you think it is a good enough solution to use this > > systeme in production? > > We've had reports from various Large Corporate Users that lowering the > tag depth to 32 results in satisfactory stability for them. YMMV. > > > Is there any other test, I could do help debug the problem or the > > stability of the system? > > Unfortunately, we need traces from a very high resolution SCSI analyzer > to capture the events that lead to the problem. If you have the > equipment, we would gladly take any information you can capture with it. > > > Since, the tag depth gets reset every reboot, is there some where that > > I fix it to 32. maybe a sysctl variable? Or should I put in the > > rc.conf if so how? > > You can create a small script to run the command and place it in > /usr/local/etc/rc.d. > > > Thank you again. > > Hopefully we'll get this solved soon =3D-) > > Scott > > > S=E9bastien > > > > Scott Long wrote: > >> Sebastien G wrote: > >>> Thank you. > >>> > >>> The firmware revision of the drive is 0003. Do you know if it is > >>> the drive or the controlor the problem? > >>> Could it be the motherboard or 32bit pci slot? Can I do something to > >>> help out resolve the bug? but I'm not an enginneer... ;-) > >>> > >>> S=E9bastien > >> > >> One thing that might help the problem for you is to lower the tag depth > >> of the drive to 32. Use the 'camcontrol tags' command to do this. > >> > >> Scott > >> > >>> Scott Long wrote: > >>>> Hi, > >>>> > >>>> We are aware of this problem and are working on a fix. > >>>> Unfortunately it > >>>> has proven to be very hard to track down. Any idea what firmware > >>>> revision is on that drive? > >>>> > >>>> Scott > >>>> > >>>> Sebastien G wrote: > >>>>> Hi, > >>>>> > >>>>> I just got a new Adaptec 29320 with a new 18 gig 15K seagate > >>>>> ST318453LW. I thought this would make a nice web server! I install > >>>>> Freebsd 4.8 rc1 so far everything is ok then while installing mys= ql > >>>>> from the port tree the hard drive & controlor crash... with the > >>>>> kind of > >>>>> error. > >>>>> [snip card dump/dmesg WHY] [Late I know but...]=20 Looks like this is still broken. I have Adaptec 29320 plus four Seagate Cheetah ST336607LW (firmware 0006 = =3D=20 latest). 4_8 dies during the SCSI neg, 20030731-JPSNAP boots okay. This is a=20 replacement setup for my works firewall. It dies during restore. Tags level= =20 was 33 post card dump so the camcontrol "fix" will probably work. Don't have an analyzer, but I can do some of the legwork if someone will "h= old=20 my hand". Pending MFCs? The existing drives are running on borrowed time so I'll have to put the ne= w=20 drives into production next week. CC me if I can help. =2D-=20 ian j hart Quoth the raven, bite me! Salem Saberhagen (Episode LXXXI: The Phantom Menace)