[Top] [All Lists]

Re: bad superblock after lvextend

To: Eric Sandeen <sandeen@xxxxxxx>
Subject: Re: bad superblock after lvextend
From: "Frank J. Buchholz" <frankb@xxxxxxxxx>
Date: Tue, 30 Nov 2004 07:41:24 -0500
Cc: linux-xfs@xxxxxxxxxxx
In-reply-to: <41AA90A2.7030007@xxxxxxx>
References: <41A35FAA.5000404@xxxxxxxxx> <41AA001A.5070304@xxxxxxxxx> <41AA90A2.7030007@xxxxxxx>
Sender: linux-xfs-bounce@xxxxxxxxxxx
User-agent: Mozilla Thunderbird 0.9 (Windows/20041103)
Thanks Eric. That was it. When I had taken the LVM back to it's original size I inadvertantly subtracted too many extents. Once I made the LVM the corrent number of extents, I was able to read the filesystem. Phew.

Big thanks!

Eric Sandeen wrote:

Frank J. Buchholz wrote:

Is it possible to repair this problem either through LVM or XFS? I noticed there are a number of achieved .vg files in /etc/lvm/archive, is it possible to restore LVM from one of these? Or is it possible to rebuild the superblock?

Are you sure your lvm volume is reassembled properly... after repair finds the first superblock, it checks to be sure it looks fairly sane, then compares it against a number of other superblocks, checking for consistency. Here it looks like it's trying to check a superblock that should be towards the end of the disk - looks like still a bit under 2T - and it's failing because it can't seek out to that location (not an xfs problem per se, this is a seek() system call that fails).

what does xfs_info on the filesystem say - I'd double check the geometry /size it reports and compare it with the size of your volume....


<Prev in Thread] Current Thread [Next in Thread>