From mboxrd@z Thu Jan 1 00:00:00 1970 From: Theodore Ts'o Subject: Re: 2nd Attempt - FSCK Errors Date: Fri, 3 May 2013 09:14:21 -0400 Message-ID: <20130503131421.GC32297@thunk.org> References: <007b01ce45ab$0ffc24f0$2ff46ed0$@ntlworld.com> <000f01ce47ec$ca7767c0$5f663740$@ntlworld.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: 'Andreas Dilger' , linux-ext4@vger.kernel.org To: Stephen Elliott Return-path: Received: from li9-11.members.linode.com ([67.18.176.11]:46116 "EHLO imap.thunk.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1760356Ab3ECNOZ (ORCPT ); Fri, 3 May 2013 09:14:25 -0400 Content-Disposition: inline In-Reply-To: <000f01ce47ec$ca7767c0$5f663740$@ntlworld.com> Sender: linux-ext4-owner@vger.kernel.org List-ID: What you've shown us makes me suspicious about whether the hardware device is sane or not. In the previous e2fsck run, it set i_size to a non-zero value. Yet when debugfs tries to read the same inode, it's now seeing all zero's. So that implies the disk (or software raid device; you haven't been clear what the underlying storage is for this file system) is not returning the same information for a particular block as was previously written. If the underlying block device is not stable, there really is nothing for e2fsck to do. You might want to check /var/log/messages for any error messages relating to the underlying storage device(s). If you're seeing I/O errors in the log files, that would be another hint. At this point, my recommendation to you is to find a separate disk (or RAID array if necessary) which is as big as the underlying disk, and do an image copy (via dd or ddrescue) to a known-good storage device, and then retry the e2fsck on this copy of the file system. Regards, - Ted