XP guest runs on XP host but crashes on linux host

[MaxMaximus]

Hello,

I created a Windows XP SP3 guest on my XP SP3 Notebook, VBox 4.0.4. It works fine.
When I copy the whole machine to my Linux server (VBox 4.0.4), it runs in to the endless loop described in several threads here.
I played around with nearly all parameters (IOAPIC, VT-x, storage controller). Only thing that changed: turning on VT-x caused vm not to restart but to hang. So I set it to off again and the restart loop was back. Ex/import also doesn't help.

Does anybody knows if there are differences between the Windows and Linux implementation of VBox that may cause those problems. Isn't it the big advantage of virtualization to run the same configured guest on different hosts?
It took me some time to set up the guest on my notebook before I roll it out on the server. Now I'm afraid that when I setup a new guest on server it will not run on my notebook and vice versa.
And, more important: Which parameter could be respnsible for that.

Main configuration items of my vm:

Page Fusion: off
HPET: off
Chipset: ich9
Firmware: BIOS
Number of CPUs: 1
Synthetic Cpu: off
CPUID overrides: None
ACPI: on
IOAPIC: on
PAE: off
Hardw. virt.ext: off
Hardw. virt.ext exclusive: off
Nested Paging: on
Large Pages: off
VT-x VPID: off
Storage Controller Name (0): IDE-Controller
Storage Controller Type (0): PIIX4
Storage Controller Instance Number (0): 0
Storage Controller Max Port Count (0): 2
Storage Controller Port Count (0): 2
Storage Controller Bootable (0): on

Thanks in advance.

[mpack]

It is unlikely that the BSOD and reset is caused by differences in the Linux vs Windows hosted versions of VirtualBox. This kind of problem is usually caused by hardware differences in the VM settings - I assume you created the VM separately on each host and only move the VDI between them? Another possibility is that the CPU is different on the two hosts, but while I can see that might cause a reactivation, I can't see it causing a crash.

[Sasquatch]

Another possibility is that the CPU is different on the two hosts, but while I can see that might cause a reactivation, I can't see it causing a crash.

It can cause a crash if he's experiencing the intelppm.sys BSOD that shows up when moving from an Intel CPU to an AMD CPU. Without the actual BSOD, we can't tell what the cause is.

[mpack]

It can cause a crash if he's experiencing the intelppm.sys BSOD that shows up when moving from an Intel CPU to an AMD CPU

True, good point.

[MaxMaximus]

Thanks for the answers.

I copy the whole machine (.vbox, .vmdk) between the two hosts. So it's the same hardware definition. That's the point I really don't understand.

CPU: Core 2 Duo vs. Core 2 Quad

I will take a snapshot of the BSOD tomorrow when I am back at office.

[mpack]

I copy the whole machine (.vbox, .vmdk) between the two hosts.

You have to take care when copying the xml control files between hosts, especially when the hosts run different operating systems. Paths will be different, path naming conventions will be different, and the xml may otherwise refer to named host resources (e.g. "COM1") that the guest uses - and the name of that resource may be totally different, or not available at all, on a different host running a different OS.

Yes, v4 no longer puts fundamental obstacles in the way of moving entire VMs between hosts, but you still need to give it some thought, especially when the hosts are so different.