From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from relay.sgi.com (relay3.corp.sgi.com [198.149.34.15]) by oss.sgi.com (Postfix) with ESMTP id 7D5267F3F for ; Thu, 19 Dec 2013 02:02:38 -0600 (CST) Received: from cuda.sgi.com (cuda3.sgi.com [192.48.176.15]) by relay3.corp.sgi.com (Postfix) with ESMTP id 04FF8AC002 for ; Thu, 19 Dec 2013 00:02:34 -0800 (PST) Received: from dakara.d-tor.org (dakara.d-tor.org [88.198.161.242]) by cuda.sgi.com with ESMTP id NzhglJpLRB4up2sG for ; Thu, 19 Dec 2013 00:02:32 -0800 (PST) Message-ID: <52B2A807.5000608@disconnect.de> Date: Thu, 19 Dec 2013 09:02:15 +0100 From: arno MIME-Version: 1.0 Subject: Re: [XFS BUG] xfs_growfs: XFS_IOC_FSGROWFSDATA xfsctl failed: Structure needs cleaning References: <52B1B89A.8030401@disconnect.de> <20131218223312.GP31386@dastard> In-Reply-To: <20131218223312.GP31386@dastard> 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: Dave Chinner Cc: linux-kernel@vger.kernel.org, xfs@oss.sgi.com On 18-Dec-13 11:33 PM, Dave Chinner wrote: > On Wed, Dec 18, 2013 at 04:00:42PM +0100, arno wrote: >> Hello, >> >> (Please keep me in CC, I'm not subscribed) >> >> Just got "xfs_growfs: XFS_IOC_FSGROWFSDATA xfsctl failed: Structure >> needs cleaning" when trying to grow a logical Volume. Sorry, I lost the >> xfs_growfs output. >> >> After collecting some info, I ran xfs_repair, and that seemed to fix it, >> so no harm done. After fixing, the FS had the new, increased size, if >> that matters. > > This isn't the first time you've grown the filesystem, is it? No, indeed it isn't. > On older kernels, growfs could leave trailing garbage in the new > secondary superblocks it created. This was fixed in kernel v3.8 by > commit: > > 1375cb6 xfs: growfs: don't read garbage for new secondary superblocks > > This garbage is detected by other changes made in v3.8 that verify > the secondary superblocks as they are read. > > 9802182 xfs: verify superblocks as they are read from disk > > So if you grew the filesystem on a kernel older than v3.8, growing > it again on a v3.8+ kernel will detect the superblock corruption and > throw this error. I'm pretty sure that last time I grew it, it was with a pre v3.8-kernel. > xfs_repair was recently fixed to detect the garbage in secondary > superblocks and zero it: > > cbd7508 xfs_repair: zero out unused parts of superblocks > > So what you need to do is run xfs_repair built from the development > git tree and run it on your filesystem. That will remove the garbage > from the superblocks and make this problem go away forever. Will do. Thanks! -- Regards. Arno. _______________________________________________ xfs mailing list xfs@oss.sgi.com http://oss.sgi.com/mailman/listinfo/xfs