From owner-freebsd-current Mon Jul 22 14:02:36 1996 Return-Path: owner-current Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id OAA08277 for current-outgoing; Mon, 22 Jul 1996 14:02:36 -0700 (PDT) Received: from GndRsh.aac.dev.com (GndRsh.aac.dev.com [198.145.92.241]) by freefall.freebsd.org (8.7.5/8.7.3) with SMTP id OAA08265 for ; Mon, 22 Jul 1996 14:02:30 -0700 (PDT) Received: (from rgrimes@localhost) by GndRsh.aac.dev.com (8.6.12/8.6.12) id OAA19362; Mon, 22 Jul 1996 14:02:20 -0700 From: "Rodney W. Grimes" Message-Id: <199607222102.OAA19362@GndRsh.aac.dev.com> Subject: Re: SCSI errors and repeated panics To: amir@neuron.net (Amir Y. Rosenblatt) Date: Mon, 22 Jul 1996 14:02:19 -0700 (PDT) Cc: amir@neuron.net, freebsd-current@freebsd.org In-Reply-To: <199607222049.QAA00424@prozac.neuron.net> from "Amir Y. Rosenblatt" at "Jul 22, 96 04:49:49 pm" X-Mailer: ELM [version 2.4ME+ PL11 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-current@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > I see justin replied about a kernel bug, but thought I might mention that > > you did not complete the scsi bus termination description: > > OK -- I re-supped last night and recompiled my kernel. > > > The upper 8 bits are terminated on the 2940UW, if not you have > > a miss termination on the data<7-15> signals of the scsi bus, > > please check your SCSI-Select BIOS settings. > > I'm not entirely sure what you mean. I mean that if you go into the SCSI-Select BIOS (Hit ^A during system Power On Self Test (POST) when the Adaptec bios mentions it) and check the settings of the on board scsi termination. You will find 3 or 4 options: a) Automatic (I recommend against this setting, the detection logic may fail under certain conditions.) b) High/Low Enabled (ie, all 16 bits of the scsi bus are terminated. c) High Enabled (ie, the upper 8 bits of the scsi bus are terminated) d) Disabled. In your configuration it _must_ be set to c) above... see the diagrams in the Adaptec manual.... > I've had devices on boith the 8 and > 16 bit internal connectors on the card since the first day I brought the > machine online back in February. The only change I've made to the setup > is that originally, the 4 gig HP C1533 DAT was at the end of the 8-bit > chain and now it's in the middle, with the NEC CD-ROM drive at the end > (the termination on both set appropriately as far as I could tell from > the manuals). Humm... I suspect that the HP C1533 DAT drive has active terminators, ie the resistor value will be 110 ohms, and the NEC CD-ROM is using passive termination, ie the resitor value will be 220/330 ohm. I highly recommend (and the SCSI-II spec just recommends) that you use active termination with Fast SCSI-II. > From the outset I'd adjusted the 2940UW's built-in > termination to match this configuration as specified in the manual. > Furthur advice or clarification would be greatly appreciated, as it > started giving me the same errors again a short while ago (I called the > guy who was sitting at the console and had him bring prozac [the machine] > down for 15 minutes rather than letting it panic and crash). Thanks in > advance, > > -Amir -- Rod Grimes rgrimes@gndrsh.aac.dev.com Accurate Automation Company Reliable computers for FreeBSD