[Top] [All Lists]

Re: Oops during umount of r/o file system

To: Stuart Rowan <strr-debian@xxxxxxxxxxxxxxxxxx>
Subject: Re: Oops during umount of r/o file system
From: Christoph Hellwig <hch@xxxxxxxxxxxxx>
Date: Sun, 15 Feb 2009 14:18:16 -0500
Cc: xfs@xxxxxxxxxxx
In-reply-to: <499540B1.2080100@xxxxxxxxxxxxxxxxxx>
References: <499540B1.2080100@xxxxxxxxxxxxxxxxxx>
User-agent: Mutt/1.5.18 (2008-05-17)
On Fri, Feb 13, 2009 at 09:43:13AM +0000, Stuart Rowan wrote:
> I fetched and installed the linux-image-2.6.26-bpo.1-amd64 package from  
> backports.org.
> This is the Debian lenny kernel, but backported to etch. It looks to be  
> and a few patches. I realise the "and a few patches" sounds bad  
> but usually with Debian they are minor things like adding device ids.
> uname -a:
> Linux tigris 2.6.26-bpo.1-amd64 #1 SMP Wed Dec 17 12:18:36 UTC 2008  
> x86_64 GNU/Linux
> We do the following as part of our backup script:
> mkdir -p /tmp/dev/data/home-xfs; /usr/sbin/xfs_freeze -f /home ;  
> /sbin/lvcreate -s -L 20G -n snap-shot /dev/data/home-xfs ;  
> /usr/sbin/xfs_freeze -u /home && mount -o nouuid,ro /dev/data/snap-shot  
> /tmp/dev/data/home-xfs
> (rsync the snapshot offsite)
> umount /tmp/dev/data/home-xfs ; /sbin/lvremove -f /dev/data/snap-shot
> The umount, intermittently, of this ro filesystem hangs until a reboot  
> as below.

Any chance you could capture the output of dmesg after an

        echo t > /proc/sysrq-trigger

to see where it hangs exactly?

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