[Fedora-xen] XEN-3.0.3 on RHEL-5, fully virtualized guest of RH72 won't start after install finishes.
Daniel P. Berrange
berrange at redhat.com
Tue Jun 12 14:15:15 UTC 2007
On Mon, Jun 11, 2007 at 05:51:17PM -0400, Ben Russo wrote:
> I have many boxes running rh-7.2
> (these are stripped down, minimal installs, for appliances on a private
> network that isn't routable) I want to run VIRTUAL images of the
> same setup for testing purposes.
>
> I read the Fedora 6 XEN getting started guide and skimmed through the
> FAQ. I browsed through this mailing list to see if I could find my
> problem here as well. No luck.
>
> I have a Dell-2950 with Quad Core XEON processors, PAE, VMX, and it is
> enabled in BIOS. 8GB RAM, 2GB assigned to Dom0
>
> Booted into RHEL-5 Xen kernel, installed all the latest patches as of
> 6/11/2007, and rebooted.
>
> Then Started Virtual Machine Manager GUI
> Selected "New..." from File menu and followed the wizard.
> then installed FULLY VIRTUALIZED
> RH-72 from original Red Hat 7.2 CD's
> this went fine with the
> Virtual Machine Manager GUI "New Machine" wizard.
> The only non-usual thing I did for an RH72 install was
> to give it the kernel boot option mem=500M
>
> After the install completed, the DomU guest said it was now going to
> reboot and I should remove any CD's.
>
> Then the DomU disappeared.
>
> It took me forever to figure out that I couldn't use the virt-manager
> GUI to restore a DomU configuration file. There really should be
> a more intuitive way for this to work. I am not a dummy, and it took
> me forever to figure out how to start the DomU again.
The virt-manager release in Fedora 7 can now start existing inactive
guests. We recently pushed an update to Fedora 6 updates-testing
which can do the same.
> Now that I figured out that I have to go to a command line prompt and
> type "xm create -c /etc/xen/rh72-test",
That is no longer neccessary. Virt-manager can do that. You can also
do it with 'virsh' which is the preferred tool for command line usage
virsh start rh72-test
> Unable to handle kernel paging request at virtual address ffffffff
> printing eip:
> c0204a00
> *pde = 00001063
> *pte = 00000000
> Oops: 0000
> CPU: 0
> EIP: 0010:[<c0204a00>]
> EFLAGS: 00010202
> eax: fffffff ebx: 00000000 ecx: 00000004 edx: 00000004
> esi: fffffff edi: c0229972 ebp: 0008e000 esp: c0257fb4
> ds: 0018 es: 0018 ss: 0018
> Process swapper (pid: 0, stackpage=c0257000)
> Stack: c026d6e8 0000000 c024....
> ...
> ...
> Call Trace: [<....
>
> Code: f3 a6 ...
> <0>Kernel panic: Attempted to kill the idle task!
Please file BZ reports for any kernel panic issues - mailing lists
are a very bad way to track kernel bugs.
Dan.
--
|=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=|
|=- Perl modules: http://search.cpan.org/~danberr/ -=|
|=- Projects: http://freshmeat.net/~danielpb/ -=|
|=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=|
More information about the Fedora-xen
mailing list