From owner-aic7xxx Sat Oct 3 23:44:45 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id XAA21956 for aic7xxx-outgoing; Sat, 3 Oct 1998 23:44:45 -0700 (PDT) (envelope-from owner-aic7xxx@FreeBSD.ORG) Received: from smtp2.andrew.cmu.edu (SMTP2.ANDREW.CMU.EDU [128.2.10.82]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id XAA21922 for ; Sat, 3 Oct 1998 23:44:24 -0700 (PDT) (envelope-from gobbel@andrew.cmu.edu) Received: from gigan (CREAMOFWHEAT-49.SLIP.ANDREW.CMU.EDU [128.2.120.49]) by smtp2.andrew.cmu.edu (8.8.5/8.8.2) with SMTP id CAA22732; Sun, 4 Oct 1998 02:43:58 -0400 (EDT) Date: Sun, 4 Oct 1998 02:43:58 -0400 (EDT) From: Randy Gobbel Reply-To: Randy Gobbel Subject: How to erase a defect list? To: aic7xxx@FreeBSD.ORG cc: Randy Gobbel Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Sender: owner-aic7xxx@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Apologies for continuing this somewhat off-topic saga, but this list is the best source of expertise on SCSI devices I know of, and I'm still pretty much in panic mode. Context: I've been getting medium errors on my hard drive, suspected the aic7xxx driver but have since been convinced the disk is actually failing. Currently trying to save my data before the whole thing dies completely. I called Quantum, and after a good bit of hassle got them to send me a "new" disk before I send them the old one, so I can copy the data to the new one. The first idiot^H^H^H^H^Hperson I talked to couldn't find the serial number of my disk in his database, so wanted me to send them the disk so they could "verify the serial number" to see if the disk was covered under warranty, and they'd send me a new one. Of course, they want $840 for an obsolete replacement disk if my disk is not covered. No thanks. I finally got a supervisor who found my disk's serial number and told me it was covered. They sent me a "new" disk, but: it's Rev D, whereas my "old" disk is Rev F, and it has firmware L912, which is known to be buggy, while the most recent firmware is L915. Not a good start. I was not able to make it through a badblocks scan on the disk without causing the aic7xxx driver to go into an (apparently) infinite timeout/abort/reset loop (with pre12--I am now running pre15 but haven't had another chance to try badblocks). Then things really went downhill, and here's where I am badly in need of some help: I ran badblocks and mistakenly pointed it at /dev/sdb, rather than /dev/sdb4. This caused very strange things to happen--the whole system blocked for quite a long time. I was finally able to abort the badblocks run, but it has done something *VERY* bad to the disk. I am now unable to access either the primary or grown defects lists. I have tried to format the disk using a) BIOS, b) the "DSP" DOS program from Quantum, and c) the scsiformat command. The only one of these that I've been able to get anything out of is DSP. It starts formatting, but the disk acts very strange indeed: it thrashes wildly during the format operation, and the format goes so slowly that I calculated that it would take about 3 WEEKS or so to finish formatting. It looks like badblocks has somehow smashed the defect and sector remapping info, so that the disk now has a rather surreal idea of what logical sector goes with what physical sector. So: HOW CAN I REINITIALIZE THIS INFORMATION???? I would like to erase the grown defect list, and if possible return the primary defect list to its pristine state. This disk has not quite achieved "brick" status, but is now perilously close. This whole thing is just unbelievably miserable. After far too much hassle to start with, Quantum sends me a replacement unit that may not be acceptable, and in the process of trying to verify this bogus replacement, the damn thing gets so screwed up I can't even format it. Any suggestions appreciated, needless to say. -Randy To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-aic7xxx" in the body of the message