Pkg xen devel - Mar 2010 - Bug#573210: xen-hypervisor-3.2-1-amd64: Xen domU sometimes hogs CPU and doesn't respond

Package: xen-hypervisor-3.2-1-amd64
Version: 3.2.1-2
Severity: normal

*** Please type your report below this line ***
Hi,

Recently I've experienced some crashes of Xen domU (also debian Lenny, 
running same kernel as dom0 (linux-image-2.6.26-xen-amd64)).
At xm top that domain appears as running, but hogs all CPUs, because 
this problem has appeared on two identic servers (Supermicro X7SBi, QC 
Xeon, 8G DDR II). I think that problem cannot be in hardware. We were 
running this configuration on some other servers with core2 duo 
processor (only dualcore) and there this problem never appeared. I've 
saved a core-dump of affected domU, but it's really big(1,1GB), so I'll 
not attach it to bugreport, but it is available for debugging.

Thanks for potential interest in debugging.

Ond?ej Kunc

-- System Information:
Debian Release: 5.0.4
  APT prefers stable
  APT policy: (500, 'stable')
Architecture: amd64 (x86_64)

Kernel: Linux 2.6.26-2-xen-amd64 (SMP w/4 CPU cores)
Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/bash

xen-hypervisor-3.2-1-amd64 depends on no packages.

Versions of packages xen-hypervisor-3.2-1-amd64 recommends:
ii  xen-utils-3.2-1               3.2.1-2    XEN administrative tools

Versions of packages xen-hypervisor-3.2-1-amd64 suggests:
pn  xen-docs-3.2                  <none>     (no description available)

-- no debconf information

On Tue, Mar 09, 2010 at 08:18:54PM +0100, Ondrej Kunc
wrote:
> Package: xen-hypervisor-3.2-1-amd64
> Version: 3.2.1-2
> Severity: normal
>
> *** Please type your report below this line ***
> Hi,
>
> Recently I've experienced some crashes of Xen domU (also debian Lenny,
> running same kernel as dom0 (linux-image-2.6.26-xen-amd64)).
> At xm top that domain appears as running, but hogs all CPUs, because  
> this problem has appeared on two identic servers (Supermicro X7SBi, QC  
> Xeon, 8G DDR II). I think that problem cannot be in hardware. We were  
> running this configuration on some other servers with core2 duo  
> processor (only dualcore) and there this problem never appeared. I've  
> saved a core-dump of affected domU, but it's really big(1,1GB), so
I'll
> not attach it to bugreport, but it is available for debugging.
>
> Thanks for potential interest in debugging.
>
Hello,

It's this problem:
http://moblog.wiredwings.com/index.php?url=archives/20090227/Lennys-Xen-Kernel-2.6.26-Causes-DomU-Freezes.html

I tried getting debug info (stack traces) of the crashed guest here:
http://lists.xensource.com/archives/html/xen-devel/2010-03/msg00281.html

but it'd need more investigations and actual debugging..

-- Pasi