From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Sun, 19 Aug 2001 09:38:04 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Sun, 19 Aug 2001 09:37:55 -0400 Received: from probity.mcc.ac.uk ([130.88.200.94]:65036 "EHLO probity.mcc.ac.uk") by vger.kernel.org with ESMTP id ; Sun, 19 Aug 2001 09:37:49 -0400 Date: Sun, 19 Aug 2001 14:38:01 +0100 From: John Levon To: linux-kernel@vger.kernel.org Subject: Nasty oops with 2.4.8 Message-ID: <20010819143801.A90831@compsoc.man.ac.uk> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.3.19i X-Url: http://www.movement.uklinux.net/ X-Record: 0898 Dave - Brack Dragon X-Toppers: N/A Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org I got an oops yesterday just after loggin in. The machine was idle. Unfortunately the problem trashed a number of files on the root partition, including all traces of the (decoded) oops. I do remember it was a pointer dereference fault in sync_unlocked_inodes(), not a NULL pointer. The machine passes memtest86 without a hitch, has been stable with 2.4.3 and earlier for ages. ext2 /, UP Celery 600 Interestingly during the fsck an fsck.ext2 bug was uncovered where inode_link_info was 0 but inode.i_links.count was 1 (no, I definitely wasn't fscking a rw / ;). Perhaps that is fixed in 1.19 though, I was using 1.18 hope this helps someone debug this rather nasty FS corruption ... john