Re: reboot after "scrubbing free ram"

["Bastian Mäuser" <mephisto@mephis.to>]

On 02/09/2011 12:12 PM, Ian Campbell wrote:
> On Wed, 2011-02-09 at 10:16 +0000, Bastian Mäuser wrote: 
>> Hello,
>>
>> I have some diagnostic Info for you. Tried to get xen-4.0.1 working on
>> another node of my cluster. Only difference: Newer revision of the
>> Hardware. Exactly same Software Stack.
>>
>> Stack: Ubuntu Lucid 64bit, xen-4.0.1 from Source, selfbuilt
>> linux-2.6.32.27 from xen/stable-2.6.32.x
>>
>> Nodes working: HP DL380G5 16GB (Intel X5460)
>>
>> Node not working: HP DL380G6 16GB (Intel X5550)
>>
>> Effect: Xen starts but dom0 Crashes immidiately after I see the message
>> "Scrubbing free Ram". I attached Screenshot of the Management Processors
>> Serial Port.
>>
>> I used the same dom0 kernel deb that successfully runs on other Nodes
>> (as well on xen-4.0.1).
>>
>> Any ideas? I'm really stuck on this. I use XEN since 5+ years, but I
>> never had that one..
> Please can you convert the faulting RIP address to a line of code. e.g.
> using
> $ gdb vmlinux
> (gdb) list *0xffff.....
I need some more Info on that. GDB'ing the Kernel Image won't work right?
> Please see also "Are there more debugging options I could enable to
> troubleshoot problems with Xen and/or dom0 kernel?"
> http://wiki.xen.org/xenwiki/XenParavirtOps for other debug options to
> try.

This is my Debugging:

  multiboot /boot/xen-4.0.1.gz dom0_mem=1024M dummy=dummy loglvl=all
guest_loglvl=all sync_console console_to_ring com1=115200,8n1
console=com1 lapic=debug apic_verbosity=debug apic=debug iommu=off
  module /boot/vmlinuz-2.6.32.27 dummy=dummy root=/dev/cciss/c0d0p1 ro
console=hvc0 earlyprintk=xen nomodeset initcall_debug debug loglevel=10
  module /boot/initrd.img-2.6.32.27


> Limiting the amount of memory given to dom0 (using the dom0_mem
> hypervisor option) would be interesting in case this is the same issue
> as Debian bug #606964.
As you can see i always allocate 1gb to dom0..

Bastian