From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Mon, 2 Apr 2001 14:13:41 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Mon, 2 Apr 2001 14:13:30 -0400 Received: from ns0.petreley.net ([64.170.109.178]:13715 "EHLO petreley.com") by vger.kernel.org with ESMTP id ; Mon, 2 Apr 2001 14:13:16 -0400 Date: Mon, 2 Apr 2001 11:12:29 -0700 From: Nicholas Petreley To: Robert NEDBAL Cc: linux-kernel@vger.kernel.org, ReiserFS Mailing List Subject: Re: ReiserFS - corrupted files (2.4.3) Message-ID: <20010402111229.B482@petreley.com> In-Reply-To: <20010402085448.A482@petreley.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.3.16i In-Reply-To: ; from R.Nedbal@sh.cvut.cz on Mon, Apr 02, 2001 at 07:27:41PM +0200 Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Robert, That output below is part of the results from reiserfsck /dev/hde5 (obviously /dev/hde5 is the partition that I checked). I'm using reiserfsprogs version 3.x.0i. You can run reiserfsck without any arguments and it can't do any harm to your partition - it's read-only. I tried "reiserfsck -x /dev/hde5" but that didn't fix anything. (the -x switch is supposed to "fix fixable without having to use --rebuild-tree). But reiserfsck --rebuild-tree /dev/hde5 fixed the bitmap problem in this case, and it didn't cause any new problems. Nevertheless, I'm using reiserfs on all my machines, so I would be glad to help debug the problems asap - I'm worried about losing a whole partition here and there eventually. I'd go back to kernel 2.2.18, which is supposedly very stable, but I can't. It doesn't work for me on reiserfs partitions created with 2.4.x. I don't know where to start with debugging, either, but I'm willing to take instructions. -Nick * Robert NEDBAL (R.Nedbal@sh.cvut.cz) [010402 10:28]: > Hello, > thank you for your fast reply. > > On Mon, 2 Apr 2001, Nicholas Petreley wrote: > > Comparing bitmaps..free block count 613785 mismatches with > > a correct one 613799. > > byte 114154: bm1: ffffffff bm2 1f > > byte 114155: bm1: ffffffff bm2 fffffffc > > byte 118307: bm1: ffffffff bm2 ffffffef > > byte 118309: bm1: ffffffff bm2 3f > > byte 118310: bm1: ffffffff bm2 fffffffc > > byte 118312: bm1: ffffffff bm2 ffffff87 > > on-disk bitmap does not match to the correct one. 6 bytes > > differ > > How you get these results? > > > > > And I found a recently updated log file with nulls in it - > > I thought that problem was fixed long ago. I've been using > > ac28 and 2.4.3 while getting these errors. The corruptions > > occur on two machines, asus a7v and asus a7v133, via686a > > and via686b. I've had reasonably good luck with > > reiserfsck, though, so it's not something to fear entirely. > > ;-) > > > > I'll try running reiserfsck as the last option. (Just before complete > reinstallation :-) ) > > > I'd be glad to help with any debugging as time permits. > > > > I'm happy that somebody is interrested in debugging. Bud I'm only C/C++ > coder, I dont know much about kernel internals. How can I begin with > debugging? > > regards, > Robert > > > > > -- > > ********************************************************** > > Nicholas Petreley Caldera Systems - LinuxWorld/InfoWorld > > nicholas@petreley.com - http://www.petreley.com - Eph 6:12 > > ********************************************************** > > . > > > > -------------------------------------------------------------------- > Robert Nedbal - Czech Technical University in Prague, Czech Republic > email: R.Nedbal@sh.cvut.cz http://www.sh.cvut.cz/~robik/ > /* Debuggers are evil. Never ever trust them. */ > -------------------------------------------------------------------- > -- ********************************************************** Nicholas Petreley Caldera Systems - LinuxWorld/InfoWorld nicholas@petreley.com - http://www.petreley.com - Eph 6:12 ********************************************************** .