From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S264134AbTLJVYX (ORCPT ); Wed, 10 Dec 2003 16:24:23 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S264136AbTLJVYX (ORCPT ); Wed, 10 Dec 2003 16:24:23 -0500 Received: from thebsh.namesys.com ([212.16.7.65]:13477 "HELO thebsh.namesys.com") by vger.kernel.org with SMTP id S264134AbTLJVYV (ORCPT ); Wed, 10 Dec 2003 16:24:21 -0500 Message-ID: <3FD78F03.7080205@namesys.com> Date: Thu, 11 Dec 2003 00:24:19 +0300 From: Hans Reiser User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.5) Gecko/20031007 X-Accept-Language: en-us, en MIME-Version: 1.0 To: Vitaly Fertman CC: Jan De Luyck , linux-kernel@vger.kernel.org, Reiserfs developers mail-list Subject: Re: forwarded message from Jan De Luyck References: <16343.2023.525418.637117@laputa.namesys.com> <200312101604.15299.vitaly@namesys.com> <3FD77A0E.7000909@namesys.com> <200312110002.29307.vitaly@namesys.com> In-Reply-To: <200312110002.29307.vitaly@namesys.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Vitaly Fertman wrote: >On Wednesday 10 December 2003 22:54, Hans Reiser wrote: > > >>Vitaly Fertman wrote: >> >> >>>Hello, >>> >>> >>> >>>>Hello, >>>> >>>>Today I discovered this in my syslogs, after something strange >>>>happening to XFree86 (hung at startup, then dumped me back to the >>>>console) >>>> >>>>is_leaf: free space seems wrong: level=1, nr_items=41, free_space=65224 >>>>rdkey vs-5150: search_by_key: invalid format found in block 283191. Fsck? >>>>vs-13070: reiserfs_read_locked_inode: i/o failure occurred trying to find >>>>stat data of [11 12795 0x0 SD] is_leaf: free space seems wrong: level=1, >>>>nr_items=41, free_space=65224 rdkey vs-5150: search_by_key: invalid >>>>format found in block 283191. Fsck? vs-13070: >>>>reiserfs_read_locked_inode: i/o failure occurred trying to find stat >>>>data of [11 12798 0x0 SD] >>>> >>>> >>>this all about fs corruptions. fsck is needed. >>> >>> >>is this a failure due to bad sector on the drive? >> >> > >No, we return EIO in many places if some data corruption is found >even if the hardware has worked ok. A stat data has not been found >here and EIO is returned. > >-- >Thanks, >Vitaly Fertman > > > > fix the code to have a more accurate description. This is not what I would consider an IO error, and it should not describe itself to users that way. Also, please attempt to determine if this is happening on a file system that has only had a recent kernel running on it. I am concerned that we might have a bug in recent V3. -- Hans