I've got two identical servers, identically installed. One has the problem
and the other one does not. The main difference of use between the two is
that one has got only 6 VMs running, the other has got 20 which are not all
running at the same time.

I'm facing the same issue as yours on the second one which is more loaded.
I've also noticed that if on that machine, I save all states and reboot,
sometimes the problem does not occure but more often it does. But if I reboot
the host OS after shutting off all guests instead of saving states, the issue
never occures.

So I did some more testing on that. After rebooting the server with most VMs
on saved state, I get the issue: first a crash, then no network on gest OSs.
If then I shut down all Guest OSs and then turn them one by one back on,
everything works fine. And this systemmatically works.

I have no idea of why the first crash happens. I've cascaded several of
those logs to MS with no answer. But still I believe that the crash happens
when trying to restore the state. So the VM must be marked as "running"
somewhere by Virtual Server. And that state never changes unless you shut the
guest OS down. This would explain why we get that duplicate MAC address
message. Virtual Server then considering that the guest OS is already on and
using the MAC addresses allocated to it.

So for the moment being, while expecting on MS for a patch to correct the
restore state issue, shutting down the guest OSs (which forsure is a heavy
task...) is my turnaround for that problem.

Hoping this s.