From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Thu, 28 Nov 2002 20:22:29 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Thu, 28 Nov 2002 20:22:29 -0500 Received: from dial-ctb03115.webone.com.au ([210.9.243.115]:3589 "EHLO chimp.local.net") by vger.kernel.org with ESMTP id ; Thu, 28 Nov 2002 20:22:27 -0500 Message-ID: <3DE6C333.4080607@cyberone.com.au> Date: Fri, 29 Nov 2002 12:30:27 +1100 From: Nick Piggin User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.1) Gecko/20020913 Debian/1.1-1 MIME-Version: 1.0 To: Alan Cox CC: "Piggin, Nick" , "'linux-kernel@vger.kernel.org'" Subject: Re: 2.4.19-rc1,2 + ext3 data=journal: data loss on unmount References: <1038502327.10021.17.camel@irongate.swansea.linux.org.uk> Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Alan Cox wrote: >On Thu, 2002-07-18 at 07:12, Piggin, Nick wrote: > >>mount /mnt/backup >>tar cvf $FILENAME directory >>bzip2 $FILENAME >>umount /mnt/backup >> >>Upon remounting and inspection, the resulting bzip2 file is corrupted every >>time. Adding a sync after bzip2 corrects the problem. >> > >That sounds like a bug in the core code somewhere since the flush should >have happened automatically on umount. Does 2.4.19 proper show this on >your box (or 2.4.20-rc) ? [I'd suspect yes but would like to be sure] > > 2.4.20 does indeed show this bug. The following script will trigger it: #!/bin/sh mount /mnt/backup dd if=/dev/urandom of=/mnt/backup/test.bin bs=1024 count=512 bzip2 /mnt/backup/test.bin umount /mnt/backup After running: mount /mnt/backup bunzip2 /mnt/backup/test.bin.bz2 bunzip2: x is not a bzip2 file. The file is filled mostly with zeros Nick