From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from smtprelay03.ispgateway.de ([80.67.31.26]:44536 "EHLO smtprelay03.ispgateway.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751748Ab2LBJtU (ORCPT ); Sun, 2 Dec 2012 04:49:20 -0500 Received: from [91.65.33.110] (helo=server.fritz.box) by smtprelay03.ispgateway.de with esmtps (TLSv1:AES256-SHA:256) (Exim 4.68) (envelope-from ) id 1Tf62Q-0005I3-Ro for linux-btrfs@vger.kernel.org; Sun, 02 Dec 2012 10:40:38 +0100 Date: Sun, 2 Dec 2012 10:40:38 +0100 (CET) From: Timo Nentwig To: linux-btrfs@vger.kernel.org Subject: 3 root nodes, -o recovery/btrfs has no effect Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; format=flowed; charset=US-ASCII Sender: linux-btrfs-owner@vger.kernel.org List-ID: Hi there, I had remote filesystem not unmounting during shutdown so I hit the reboot button. Did so a couple of times in the past. 3.6.8-1-ARCH, btrfs-progs 0.19.20121005-4, don't remember the exact version I used when creating the filesystem (0.19 something). btrfsck looks like this: checking extents checking fs roots parent transid verify failed on 167047168 wanted 713090 found 713088 parent transid verify failed on 167047168 wanted 713090 found 713088 parent transid verify failed on 167047168 wanted 713090 found 713088 parent transid verify failed on 167047168 wanted 713090 found 713088 Ignoring transid failure parent transid verify failed on 167047168 wanted 713090 found 713088 Ignoring transid failure (repeating ~1800 times, always with the same IDs) root 5 inode 8156558 errors 400 root 5 inode 9930070 errors 800 root 5 inode 9930168 errors 800 Tried -o recovery, mounts instantly and apparently doesn't change anything; tried btrfs --repair, doesn't appear to change anything as well. Did mkfs.btrfs in order to restore from backup but mounting failed (broken superblock something). How can a freshly mkfs'ed fs not mount?! I restored a dd dump I took beforehand, mounts, boots but is broken. Any suggestions? thx tcn BTW what do multiple root nodes mean actually? That I do have 3 different trees with individual nodes pointing to different versions of data?