From owner-freebsd-bugs Thu May 25 19:10:32 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id TAA29443 for bugs-outgoing; Thu, 25 May 1995 19:10:32 -0700 Received: from hutcs.cs.hut.fi (root@hutcs.cs.hut.fi [130.233.192.2]) by freefall.cdrom.com (8.6.10/8.6.6) with SMTP id TAA29430 for ; Thu, 25 May 1995 19:10:26 -0700 Received: from shadows.cs.hut.fi by hutcs.cs.hut.fi with SMTP id AA14727 (5.65c8/HUTCS-S 1.4 for ); Fri, 26 May 1995 05:10:13 +0300 From: Heikki Suonsivu Received: (hsu@localhost) by shadows.cs.hut.fi (8.6.10/8.6.10) id FAA12487; Fri, 26 May 1995 05:10:12 +0300 Date: Fri, 26 May 1995 05:10:12 +0300 Message-Id: <199505260210.FAA12487@shadows.cs.hut.fi> To: davidg@Root.COM Cc: freebsd-bugs@freefall.cdrom.com In-Reply-To: David Greenman's message of 22 May 1995 07:34:28 +0300 Subject: Re: MAJOR problem with FreeBSD-2.0-RELEASE Organization: Helsinki University of Technology, Otaniemi, Finland Sender: bugs-owner@FreeBSD.org Precedence: bulk I've personally had bad experiance with Buslogic tech support for the last several months. I've made multiple calls to them, spent more than $50 in long distance waiting on hold, and when I finally gave up waiting on hold and left Buslogic may be loosing it as cheap NCR SCSI cards have taken over the market. Thus they might be downsizing and concentrate on something else. where the card's BIOS would sometimes hang just before starting the device probe on the SCSI bus - it made rebooting wcarchive extremely risky; more than 2/3rds of the time the machine would hang rather than boot. It wasn't a I'm currently having serious trouble with disks on one of our machines. This had the same symptoms, the system paniced after SCSI locking up and it wouldn't find the disks after boot. So, I removed BT and replaced it with a NCR controller. Result is that it still gets SCSI lockups, but now it seems to survive the panic as the controller BIOS apparently does reset the bus which BT isn't apparently doing. The is why FreeBSD SCSI driver doesn't survive this? It probably should do a SCSI reset somewhere, but instead it just returns I/O error for all subsequent accesses until the system panics or someone reboots. We have got three disks on the system, 1G seagate hawk, 2G seagate barracuda and IBM 0662, but the troublemaker is still unclear; the errors turn out from any of the disks in the system, not always the same one, but all accesses to the disk which first generated an error fail. Everything around the disks, including cables have been changed several times, so it is either the disks or the software. -- Heikki Suonsivu, T{ysikuu 10 C 83/02210 Espoo/FINLAND, hsu@cs.hut.fi home +358-0-8031121 work -4513377 fax -4555276 riippu SN