Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 13 Feb 2012 15:01:08 -0800
From:      Devin Teske <devin.teske@fisglobal.com>
To:        Doug Barton <dougb@FreeBSD.org>
Cc:        freebsd-fs@FreeBSD.org, freebsd-stable@FreeBSD.org, "Robison, Dave" <Dave.Robison@fisglobal.com>
Subject:   RE: Re: Why won't 8.2 umount -f?
Message-ID:  <08c101cceaa3$5ff9d880$1fed8980$@fisglobal.com>
In-Reply-To: <4F3993C9.1040506@fisglobal.com>
References:  <4F3993C9.1040506@fisglobal.com>

next in thread | previous in thread | raw e-mail | index | archive | help
> -------- Original Message -------- 
> Subject: Re: Why won't 8.2 umount -f?
> Date: Mon, 13 Feb 2012 13:20:55 -0800
> From: Doug Barton <dougb@FreeBSD.org>
> Organization: http://SupersetSolutions.com/
> To: <freebsd-stable@FreeBSD.org>
> CC: <freebsd-fs@FreeBSD.org>
> 
Cross posting? (gasp!)

> On 02/13/2012 13:02, Doug Barton wrote:
> > Is there some magic I'm missing to convince an 8.2 system to umount -f?
> > I had an NFS server crash, so I'm trying to get the mounts updated. All
> > of the 7.x systems happily did 'umount -f', but the 8.x systems (mostly
> > 8.2-pN) are just hanging forever.
> 
> ... and it gets worse. I just 'shutdown -r now'ed one of my
> less-critical 8.2 systems, and it hung for several minutes after "All
> buffers synced." After a power cycle it came back, but the buffers
> weren't actually synced because it's still fsck'ing some pretty large
> file systems.
> 
> What the heck?

We've noticed this behavior too (on 8.1-RELEASE-p6).

The work-around (to avoid long fsck) is to:

1. Drop to the kernel debugger (Ctrl+Alt+ESC -- requires DDB to be enabled in
kernel)

2. Type: call boot(0)

This causes the system to attempt a second sync of the buffers. This second
attempt ends up timing out but succeeds in resetting the CPU (after 3x 60s
timeouts).

The price (waiting 180s before cpu_reset() is called) can be well worth it
(avoiding multi-hour fsck) because the disks will be marked clean.

For us, this is a serious issue and like Doug, we too exclaim "what the heck?"
Shouldn't have to drop to kernel debugger and [redundantly] invoke boot(0) after
syncer's hang just prior to cpu_reset().
-- 
Devin

_____________
The information contained in this message is proprietary and/or confidential. If you are not the intended recipient, please: (i) delete the message and all copies; (ii) do not disclose, distribute or use the message in any manner; and (iii) notify the sender immediately. In addition, please be aware that any message addressed to our domain is subject to archiving and review by persons other than the intended recipient. Thank you.



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?08c101cceaa3$5ff9d880$1fed8980$>