From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from relay.sgi.com (relay1.corp.sgi.com [137.38.102.111]) by oss.sgi.com (Postfix) with ESMTP id 83AE67F4E for ; Fri, 4 Oct 2013 16:04:20 -0500 (CDT) Received: from cuda.sgi.com (cuda1.sgi.com [192.48.157.11]) by relay1.corp.sgi.com (Postfix) with ESMTP id 4680E8F8049 for ; Fri, 4 Oct 2013 14:04:17 -0700 (PDT) Received: from ipmail06.adl2.internode.on.net (ipmail06.adl2.internode.on.net [150.101.137.129]) by cuda.sgi.com with ESMTP id Y35gFRAMZZEEHllw for ; Fri, 04 Oct 2013 14:04:15 -0700 (PDT) Date: Sat, 5 Oct 2013 07:04:11 +1000 From: Dave Chinner Subject: Re: RESEND: xattr issue with 3.11 kernel Message-ID: <20131004210411.GJ4446@dastard> References: MIME-Version: 1.0 Content-Disposition: inline In-Reply-To: List-Id: XFS Filesystem from SGI List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Errors-To: xfs-bounces@oss.sgi.com Sender: xfs-bounces@oss.sgi.com To: David Zafman Cc: ceph-devel@vger.kernel.org, xfs@oss.sgi.com On Fri, Oct 04, 2013 at 01:48:17PM -0700, David Zafman wrote: > > <> > > This was also seen on a customized 3.10 kernel, but didn't reproduce on the Ubuntu 13.04 packaged 3.8 kernel. > > $ cat /etc/issue > Ubuntu 13.04 \n \l > $ uname -a > Linux ubuntu 3.11.0-031100-generic #201309021735 SMP Mon Sep 2 21:36:21 UTC 2013 x86_64 x86_64 x86_64 GNU/Linux > $ mount | grep /mnt > /dev/sdb on /mnt type xfs (rw) > $ ./xattr-test.sh /mnt/foo 10240 > 8 > 16 > 32 > 64 > 128 > 256 > 512 > 1024 > 2048 > setfattr: /mnt/foo: Structure needs cleaning > failure to remove user.test1557 > > ffff88003a005000: 46 c4 4f 2f 00 00 06 b5 46 c4 4f a6 00 00 06 b4 F.O/....F.O..... > ffff88003a005010: 46 c4 4f a7 00 00 06 b3 46 c4 4f a8 00 00 06 b2 F.O.....F.O..... > ffff88003a005020: 46 c4 4f a9 00 00 06 b1 46 c4 4f aa 00 00 06 b0 F.O.....F.O..... > ffff88003a005030: 46 c4 4f ab 00 00 06 af 46 c4 4f ac 00 00 06 ae F.O.....F.O..... > XFS (sdb): Internal error xfs_da_do_buf(2) at line 2694 of file /home/apw/COD/linux/fs/xfs/xfs_da_btree.c. Caller 0xffffffffa0304423 > > CPU: 0 PID: 35058 Comm: setfattr Not tainted 3.11.0-031100-generic #201309021735 > Hardware name: VMware, Inc. VMware Virtual Platform/440BX Desktop Reference Platform, BIOS 6.00 07/31/2013 > 0000000000000001 ffff8800354ab9f8 ffffffff81720fca 0000000000000007 > ffff88003a27b800 ffff8800354aba18 ffffffffa02cd27f ffffffffa0304423 > 0000000000000001 ffff8800354aba58 ffffffffa02cd2ee 0000000000000002 > Call Trace: > [] dump_stack+0x46/0x58 > [] xfs_error_report+0x3f/0x50 [xfs] > [] ? xfs_da3_node_read+0x23/0xd0 [xfs] > [] xfs_corruption_error+0x5e/0x90 [xfs] > [] xfs_da_read_buf+0x1ba/0x230 [xfs] > [] ? xfs_da3_node_read+0x23/0xd0 [xfs] > [] ? xfs_trans_log_buf+0x97/0x100 [xfs] > [] xfs_da3_node_read+0x23/0xd0 [xfs] > [] xfs_da3_blk_unlink.isra.13+0x16c/0x1e0 [xfs] > [] xfs_da3_join+0x9f/0x240 [xfs] > [] xfs_attr_node_removename+0x192/0x300 [xfs] You are probably hitting the problem fixed in commit 997def2 ("xfs: fix node forward in xfs_node_toosmall") which went into 3.12-rc3. Can you test that this fixes your problem, ad if so, we'll need to push it back to -stable kernels.. Cheers, Dave. -- Dave Chinner david@fromorbit.com _______________________________________________ xfs mailing list xfs@oss.sgi.com http://oss.sgi.com/mailman/listinfo/xfs