Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 29 Oct 2004 12:23:48 -0500
From:      "Kevin D. Kinsey, DaleCo, S.P." <kdk@daleco.biz>
To:        Charles Swiger <cswiger@mac.com>
Cc:        freebsd-questions@freebsd.org
Subject:   Re: shutdown(8): supposed to dismount all FSes?
Message-ID:  <41827CA4.4030001@daleco.biz>
In-Reply-To: <16FF66C8-29C9-11D9-9599-003065ABFD92@mac.com>
References:  <20041028185347.GA1917@zaba-s-2.hpl.hp.com> <4181C1A2.2040707@mac.com> <41822F3B.5070108@zaba.com> <16FF66C8-29C9-11D9-9599-003065ABFD92@mac.com>

next in thread | previous in thread | raw e-mail | index | archive | help
Charles Swiger wrote:

> On Oct 29, 2004, at 7:53 AM, Stefek Zaba wrote:
>
>> Chuck Swiger wrote:
>>
>>> ...I recall that there are some issues handling ext2fs filesystems 
>>> cleanly during the shutdown process, which tend to cause them to not 
>>> be unmounted properly.  It's considered a bug.
>>
>> Thanks, Chuck. When I tried searching the PRs for this, I came up 
>> blank; but your reply caused me to search harder, and indeed this is 
>> a dup of 'PR kern/56675: Syncer "giving up" on buffers and ext2 
>> filesystems'
>> which is still Open, reported 10sep03. My experience suggests it's 
>> still a problem with 5.3-RC1.
>
>
> You're certainly welcome, and I'm pleased that my memory was working 
> well enough to be helpful.  :-)
>
>> Obviously, on a "production" FreeBSD system, this bug is of minor 
>> relevance, as you'll be running UFS throughouth. But for the 
>> multibooters among us, who are looking (say) to straddle the Linux 
>> and BSD worlds, being able to mount the "old" ext2/ext3 partitions 
>> routinely is a Good Thing in easing the transition; so for such users 
>> the bug is more important.
>
>
> You might find that unmounting the ext2/ext3 filesystem by hand before 
> rebooting is an adequate workaround, since that ought to leave the 
> filesystem in a clean state.
>

Hmm, hate to come into this late, and perhaps with no good reason, but I 
have
experienced this recently with a 200GB **ufs** volume:

$uname -a
FreeBSD archangel.daleco.biz 5.3-RC1 FreeBSD 5.3-RC1 #0: Wed Oct 20 
10:10:40 CDT 2004     
root@archangel.daleco.biz:/usr/obj/usr/src/sys/GENERIC  i386

$mount
/dev/ad0s1a on / (ufs, local)
devfs on /dev (devfs, local)
/dev/ad0s1e on /usr (ufs, local, soft-updates)
/dev/ad0s1d on /var (ufs, local, soft-updates)
/dev/ad1s1d on /backup (ufs, local, soft-updates)

$sudo tail /var/log/dmesg.today
Timecounter "TSC" frequency 2404108844 Hz quality 800
Timecounters tick every 1.000 msec
ipfw2 initialized, divert enabled, rule-based forwarding disabled, 
default to deny, logging disabled
ad0: 38166MB <ST340014A/3.06> [77545/16/63] at ata0-master UDMA100
ad1: 190782MB <WDC WD2000JB-00GVA0/08.02D08> [387621/16/63] at 
ata0-slave UDMA100
Mounting root from ufs:/dev/ad0s1a
WARNING: /backup was not properly dismounted
Accounting enabled

I rebooted the box via gdm on :0:0 (ttyv0) yesterday after viewing
the initial posts of this thread.

If there is a chance I'm doing something wrong, I'd appreciate
someone with knowledge letting me know ... OTOH, if there is
a bug in 5.3 that's not been fixed (I currently am browing the
CVS logs at freebsd.org, but hardly know where to look on this
issue) related to **ufs**, perhaps we ought to ring a bell somewhere?

Kevin Kinsey
DaleCo, S.P.



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?41827CA4.4030001>