From owner-freebsd-current@FreeBSD.ORG Wed Jul 16 05:16:37 2003 Return-Path: Delivered-To: freebsd-current@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 290B837B401 for ; Wed, 16 Jul 2003 05:16:37 -0700 (PDT) Received: from spork.pantherdragon.org (spork.pantherdragon.org [206.29.168.146]) by mx1.FreeBSD.org (Postfix) with ESMTP id 7CC0F43F93 for ; Wed, 16 Jul 2003 05:16:36 -0700 (PDT) (envelope-from dmp@bitfreak.org) Received: from speck.techno.pagans (12-206-23-247.client.attbi.com [12.206.23.247]) by spork.pantherdragon.org (Postfix) with ESMTP id 07F5A2A41E for ; Wed, 16 Jul 2003 05:16:36 -0700 (PDT) Received: from speck.techno.pagans (localhost.techno.pagans [127.0.0.1]) by speck.techno.pagans (Postfix) with SMTP id F0FE15AB9 for ; Wed, 16 Jul 2003 05:16:33 -0700 (PDT) Date: Wed, 16 Jul 2003 05:16:29 -0700 From: Darren Pilgrim To: current@freebsd.org Message-Id: <20030716051629.39cce343.dmp@bitfreak.org> X-Mailer: Sylpheed version 0.8.9claws (GTK+ 1.2.10; i386-portbld-freebsd5.0) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Subject: 'Hard error reading fsbn' in 5.1 but not 5.0? X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Discussions about the use of FreeBSD-current List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 16 Jul 2003 12:16:37 -0000 Just now, my 120GB western digital, immediately upon the mounting of all filesystems during startup, made a faint "click", then my screen filled with errors of the form: ad6: hard error reading fsbn -# of -#--# error=10 That's not an exact quote, of course, since the messages scrolled by so fast I couldn't actually read parts of them. Those # signs are in place of large numbers (most were 12-18 digits). This was in 5.1-R, fresh install, in the first 15G of the disk, just after having booted into DOS to create some DOS partitions after the FreeBSD slice. I booted back into 5.0, and am presently writing across the entire disk with this command: fdisk -I /dev/ad6 && dd if=/dev/zero of=/dev/ad6s1 bs=1m At more than 33 minutes into it at 24+MB/sec, there have been no such errors and the portion of the disk that was being mounted has already been covered by zeros. After dd completes/dies, I'm going to try to reproduce the error, but I figured I'd bring this up now in case I've tripped over something someone else already knows about. I did notice that when I used fdisk(8) to look at the partition table, the slice table showed the new DOS slice starting before the end of the FreeBSD slice. Assuming that was real overlap, and not harmless output screwiness as a result of using two very different fdisks, is it possible for such corruption to cause hard read errors?