From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Tue, 25 Mar 2003 20:20:37 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Tue, 25 Mar 2003 20:20:37 -0500 Received: from adsl-67-117-70-243.dsl.sntc01.pacbell.net ([67.117.70.243]:12162 "HELO top.worldcontrol.com") by vger.kernel.org with SMTP id ; Tue, 25 Mar 2003 20:20:36 -0500 From: brian@worldcontrol.com Date: Wed, 26 Mar 2003 01:30:27 -0800 To: linux-kernel@vger.kernel.org Subject: Can't recover from broken ext3 journal Message-ID: <20030326093027.GA14108@localhost.localdomain> Mail-Followup-To: Brian Litzinger , linux-kernel@vger.kernel.org Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline X-No-Archive: yes X-Noarchive: yes User-Agent: Mutt/1.5.4i Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org I had a harddrive develop hard IO errors in the section of the drive that holds the ext3 journal. If it try to fsck the filesystem, fsck eventually terminates after trying to replay the journal. I tried to use debugfs but that fails in similar way. When I use tune2fs to try to turn off the journal, even with the -f option, it simply reports 'the needs_recovery flag is set, please run e2fsck'. I've replaced the drive and restored from backups, however, it seems rather poor design that I should lose *all* my data on the drive simply because the journal won't replay due to an disk error. The man pages for 'tune2fs' even implies you can use it for the very purpose of removing a broken journal but it doesn't work, as I reported above. I could comment out the check in the tune2fs source code, however, is there a proper way to do this? Thanks, -- Brian Litzinger