[one-users] Cold Migration not completely working. Maybe because of ACPI?

[Adnan Pasic]

Hello,
so, now I gathered some information about the workflow of the cold migration and found out that ACPI should be correctly configured for libvirt to suspend the VM and resume it afterwards.
I tried it first with a debian lenny VM, but unfortunately had no luck making it work as expected. Until before the migration I am able to ping this VM, but as soon as the migration starts I receive a "Destination Host Unreachable" and it doesn't heal itself after some time either. Strangely, "acpi", "acpid" and "acpi-support" are all installed...and "pm-utils" as well, but still no luck! 

Another strange thing is, that the command "acpi -V" gives me back following result:

No support for device type: power_supply
Cooling 0: Processor 0 of 0

I also tried it with an Ubuntu VM, but unfortunately with the same results. Could it be that my cluster-PCs just don't have the needed hardware for the ACPI-functions, which is why the VMs also aren't able to do that?

Of course I have set FEATURES = [ acpi = "yes" ] in my template...
Any ideas on where I should look next? Thanks!

Regards
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.opennebula.org/pipermail/users-opennebula.org/attachments/20110801/bc8cf397/attachment-0002.htm>