[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Xen-users] Report of "soft lockup" for a domU when paused and unpaused


  • To: xen-users@xxxxxxxxxxxxxxxxxxx
  • From: "john maclean" <jayeola@xxxxxxxxx>
  • Date: Thu, 2 Nov 2006 13:31:05 -0100
  • Delivery-date: Thu, 02 Nov 2006 13:57:24 -0800
  • Domainkey-signature: a=rsa-sha1; q=dns; c=nofws; s=beta; d=gmail.com; h=received:message-id:date:from:to:subject:in-reply-to:mime-version:content-type:content-transfer-encoding:content-disposition:references; b=SLUm5WYSh1SiDM3Y9LbB0iH2hV76Tcjfki1iTEUwd7ctGlIFQuxqy5XiGT0DVUr5sYyvZ34gqoSssM+xZMDK8kus1Tol5RmBF7edoSIa+hKrfmK0Z+Je21lhri+t6mF8o/7McsB3hgQRR3cpdstTZdjJlsYX5kv7g99wDBHfbsU=
  • List-id: Xen user discussion <xen-users.lists.xensource.com>

Hey chaps,

I'm having **more** of these soft lockup warnings. This time around I
start domUs with no network, (mainly for testing scripts). They all
get very slooooow. Take about 10 seconds to echo key presses and after
a minute they flood the terminal with the " soft lockup detected on
CPU#0!" warnings. I'd love to able to work this out myself so that I
can fix things...
[1] Would looking at /boot/System.map-2.6.16-xen help me at all?
[2]google is a very big *vserse and it's very easy to get
lost,sidetracke and distracted. What files in /var/log/can i use to
examine these errors?
[3] Anything in proc?
[4] Is this a call for *shudder* the devel list?

On 19/10/06, Tim Post <tim.post@xxxxxxxxxxxxxxx> wrote:
I've gotten that on my Desktop (which currently runs xen 3.0.2)

Dom-0 is my desktop, KDE/Gnome with no dom0_maxmem set, I just let it
balloon as I bring up guests.

My lab VM's are file backed VBD's also (including swap). I see those
messages a few times a week but have yet to pin anything of any kind of
serious consequence on them. I've just been logging when I see them and
ignoring the messages otherwise.

Since this doesn't happen on a similar machine using partition backed
VBD's for swap, my guess is it has something to do with the host having
to write to a few dom-u file systems as they're swapping too, when all
use loops it happens.

Anyone else shed some light on why? I'm not really concerned about it,
but I'd love to be able to explain it.

Best,
-Tim

On Thu, 2006-10-19 at 01:12 +0100, john maclean wrote:
> This is a message from within the domU "etch_play", based on Debian
> Etch. I have a number of foo_play templates that I experiment with.
>
> It is a file-based image filled with packages using Debian's
> debootstrap. The domU was idling away not doing anything at all and I
> ran `xm pause <domu id>` and unpaused it after five minutes or so. I'd
> love to learn *myself* how to trace/debug this error but in the
> meantime can any one tell me if this unusual or serious? Let me know
> if there's any more info required as I'd rather not flood this list
> with over enthusiastic data.
>
> etch_play:~# BUG: soft lockup detected on CPU#0!
>
> Pid: 0, comm:              swapper
> EIP: 0061:[<c01013a7>] CPU: 0
> EIP is at 0xc01013a7
>   EFLAGS: 00000246    Tainted: GF      (2.6.16-xen #1)
>   EAX: 00000000 EBX: 00000001 ECX: 00000000 EDX: 00000001
>   ESI: 00000000 EDI: c03c0000 EBP: 00000000 DS: 007b ES: 007b
>  CR0: 8005003b CR2: b7f73ef0 CR3: 003e4000 CR4: 00000640
>  [<c010301a>] xen_idle+0x4a/0xb0
>  [<c03c2260>] unknown_bootoption+0x0/0x280
>  [<c03c2260>] unknown_bootoption+0x0/0x280
>  [<c0103194>] cpu_idle+0x74/0xf0
>  [<c03c2260>] unknown_bootoption+0x0/0x280
>  [<c03c2808>] start_kernel+0x328/0x3e0
>  [<c03c2260>] unknown_bootoption+0x0/0x280
>
>




--
John Maclean  - 07739 171 531
MSc (DIC)

_______________________________________________
Xen-users mailing list
Xen-users@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-users


 


Rackspace

Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.