From: Theodore Ts'o Subject: Re: e2fsck extremly slow after: EXT4-fs.. ext4_check_descriptors: Checksum for group .. failed Date: Mon, 12 Nov 2012 11:16:46 -0500 Message-ID: <20121112161646.GF4895@thunk.org> References: <20121109000156.GQ19977@thunk.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: linux-ext4@vger.kernel.org To: "kaefert@gmail.com" Return-path: Received: from li9-11.members.linode.com ([67.18.176.11]:60670 "EHLO imap.thunk.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753687Ab2KLQQt (ORCPT ); Mon, 12 Nov 2012 11:16:49 -0500 Content-Disposition: inline In-Reply-To: Sender: linux-ext4-owner@vger.kernel.org List-ID: On Sun, Nov 11, 2012 at 07:14:40PM +0100, kaefert@gmail.com wrote: > > So it took several days, but after running it manually it finished to > run after a few days. However, It doesn't seem to get the filesystem > in a truely clean state, although it doesn't print an error (at least > not at the end), because I've ran e2fsck again on the same partition > and it found errors again. Here's the output of the last run that > completed: You said this is an external USB drive, right? How big is it? If it's affordable, something I would suggest doing is to make image copy (i.e., using dd or dd_rescue) to another external USB drive, just to rule out hardware issues. The Pass 1B/1C/1D errors, particularly if you are seeing the exact same pattern after running a full e2fsck -fy run, makes me suspicious that inode table blocks are getting written to the wrong location on disk --- and whether this is caused by the storage device failing in some strange way. Also, can you save the output of e2fsck to a file? Direct the output to a log file, so I can look at it. There are patterns of the "millions of entries of the same pattern" which you've elided which can be a hint. Also, can you disable the German translation to make it easier for me to investigate? Thanks, - Ted