Subject: Re: How long should an xfs_freeze take?
From: Steve Lord <lord@sgi.com>
To: Chris Pascoe <c.pascoe@itee.uq.edu.au>
Cc: linux-xfs@oss.sgi.com
In-Reply-To: <Pine.GSO.4.40.0202010758001.27162-100000@mango.csee.uq.edu.au>
References: <Pine.GSO.4.40.0202010758001.27162-100000@mango.csee.uq.edu.au>
Content-Type: text/plain
Content-Transfer-Encoding: 7bit
Date: 31 Jan 2002 16:07:07 -0600
Message-Id: <1012514827.26397.257.camel@jen.americas.sgi.com>
Mime-Version: 1.0
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Status: O
Content-Length: 2098
Lines: 61

On Thu, 2002-01-31 at 16:04, Chris Pascoe wrote:
> Hi Steve,
> 
> > > How repeatable is this?
> > >
> > > The problem was persistent across reboots - I rebooted to install a
> > > different LVM version at some stage (move to 1.0.2, from 1.0.1), and it
> > > occurred five times in a row after that.  I just rebooted now to say
> that it
> > > still happens - but, alas - it doesn't want to any more.  I'll try again
> at
> > > some random times throughout the day.
> 
> I can repeat this fairly reliably now - do a meaningless rsync across the
> disk (i.e. rsync itself to itself - makes it stat every file on the disk, I
> guess), then try an xfs_freeze after the rsync completes:
> 
> rsync -aH /tst1/ /tst1/
> time xfs_freeze -f /tst1
> 
> It's been hung there for ~10 minutes now, I've repeated this twice....
> rebooting - will try again...  This time it doesn't want to do it.. :(
> 
> This time I've done a rsync with real data:
> 
> testbox# time rsync -e ssh -avH --stats root@fs:/home/01/ /tst1/ ; time xfs_freeze -f /tst1  ; time xfs_freeze -u /tst1
> 
> Number of files: 602545
> Number of files transferred: 292
> Total file size: 62690390294 bytes
> Total transferred file size: 1728372390 bytes
> Literal data: 177119547 bytes
> Matched data: 1551577350 bytes
> File list size: 15575582
> Total bytes written: 4690480
> Total bytes read: 195767044
> 
> wrote 4690480 bytes  read 195767044 bytes  100808.41 bytes/sec
> total size is 62690390294  speedup is 312.74
> 
> real    33m7.243s
> user    1m34.200s
> sys     2m8.250s
> 
> I'm sitting waiting for the xfs_freeze again - so I can replicate
> the problem, yes.  It takes at least 40 minutes to complete xfs_freeze's
> after this point (i.e. you can go xfs_freeze -f ; xfs_freeze -u ;
> xfs_freeze -f ; xfs_freeze -u, and the second freeze still takes
> forever!), during which time the machine is completely unresponsive.
> 
> Chris

We are looking - or at least Eric is....

Steve

-- 

Steve Lord                                      voice: +1-651-683-3511
Principal Engineer, Filesystem Software         email: lord@sgi.com