xfs
[Top] [All Lists]

Re: permanent kernel upgrading (was: bug and fun with XFS: unable to han

To: xfs@xxxxxxxxxxx, Dave Chinner <david@xxxxxxxxxxxxx>
Subject: Re: permanent kernel upgrading (was: bug and fun with XFS: unable to handle kernel NULL pointer dereference)
From: Michael Monnerie <michael.monnerie@xxxxxxxxxxxxxxxxxxx>
Date: Mon, 26 Jul 2010 07:52:54 +0200
In-reply-to: <20100726001859.GD655@dastard>
Organization: it-management http://it-management.at
References: <201007260019.51568@xxxxxx> <20100726001859.GD655@dastard>
User-agent: KMail/1.12.4 (Linux/2.6.34.1-zmi; KDE/4.3.5; x86_64; ; )
On Montag, 26. Juli 2010 Dave Chinner wrote:
> > # mount /disks/work/
> > # umount /disks/work/
> > umount: /disks/work: device is busy.
> >         (In some cases useful info about processes that use
> >          the device is found by lsof(8) or fuser(1))
> 
> Some other process has taken a reference to the fs, I'd say.
> And if that process triggered an oops, then you'd see this.

This is definitely improbable ;-). I stopped all processes, except sshd 
and mingetty (for the console processes), and "lsof" reported no open 
files. I'd swear it must be a bug, but as you said 2.6.27 (openSUSE 
11.1) is just too old to take care.
 
> Corrupt attribute fork - matches with the oops signatures.  I'd
> definitely consider upgrading your kernel as a first step...
 
That's why I wrote "which I planned to migrate". I was just in the 
process of doing so.

This brings me to another question: openSUSE 11.1 is, despite being old, 
still maintained, so I was under the impression that I'd get updates for 
problems. But I already received "upgrade your kernel" the last time for 
my openSUSE 11.2 with it's kernel vmlinuz-2.6.31.12, which by the time 
was the latest release (in the meantime openSUSE 11.3 got released).
So is this just a generic recommendation of developers who love the 
bleeding-edge stuff and know how many problems they fixed since 2.6.27 
("yak, that old stuff, how could it ever run?"), or is it known that not 
enough fixes go into "maintained" kernels by distributions?

I have tons of servers to maintain, and can't compile kernels on my own 
anymore (I did this when I was younger), so I used openSUSE and I am in 
the process of replacing servers to SLES, which should receive much 
better services but still kernel upgrading is not supported on such 
platforms. So what is the best practice that one should choose when 
running "long time services" where you can't upgrade all the time?

Just curious. :-)

-- 
mit freundlichen Grüssen,
Michael Monnerie, Ing. BSc

it-management Internet Services
http://proteger.at [gesprochen: Prot-e-schee]
Tel: 0660 / 415 65 31

****** Aktuelles Radiointerview! ******
http://www.it-podcast.at/aktuelle-sendung.html

// Wir haben im Moment zwei Häuser zu verkaufen:
// http://zmi.at/langegg/
// http://zmi.at/haus2009/

Attachment: signature.asc
Description: This is a digitally signed message part.

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