When I attempt to Vmotion a guest from an older host to a new one I get the following error message. I know that it is related to the differences within the CPUs but I'm trying to find out how to mask the differences.
The error message is "Host CPU is incompatible with the virtual machine's requirements at the CPUID level 0x1 register 'eax'"
I've done some searching but I cannot find anything specific to eax.
I am using Intel CPU's and all guest are 32-bit.
________________________________
Jason D. Langdon
Check CPUID mask in VM settings, reset to none.
---
VMware vExpert '2009
Do you have EVC set on the cluster?
Do you have EVC set on the cluster?
No I do not. It is my understanding that I need to power off all VM's before I can enable EVC but at 260VM's, this is not an option right now.
________________________________
Jason D. Langdon
Yes, it requires the VM to be power down and change the "Edit Settings->Options->CPUID Mask->Hide NX Flag from Guest->option and you should be able to migrate. For long term solution, I would look into Enhance VMotion Compatibility if the hardwares allow to do so. I have to change the NX flag for several servers so it should work. Note: Change NX flag only require to power down the VM.
If you found this information useful, please consider awarding points for "Correct" or "Helpful". Thanks!!!
Regards,
Stefan Nguyen
VMware vExpert 2009
iGeek Systems Inc.
VMware, Citrix, Microsoft Consultant
Its a good solution but not always we have spare ESX hosts to share the loads for migration but for some folks it would be great though.
If you found this information useful, please consider awarding points for "Correct" or "Helpful". Thanks!!!
Regards,
Stefan Nguyen
VMware vExpert 2009
iGeek Systems Inc.
VMware, Citrix, Microsoft Consultant
The error message is "Host CPU is incompatible with the virtual machine's requirements at the CPUID level 0x1 register 'eax'"
Just got off the phone with VMware Support and the only way to resolve my issue is to take an outage on all VM's. Not the solution I was looking for and I hope VMware does something to address this issue in the future as restarting 250+ VM's is a lot of work with regards to ITIL standards.
________________________________
Jason D. Langdon
Sorry to hear that. Good luck.
Well, its a good standard operation checkpoint whenever you architect new solution and make sure everything is checked before deploying bunch of VMs in a cluster. Always plan and architect accordingly that's the key. Sadly, you have to do so, how would your boss respond to this 250+ VMs need to be down? If its a big cluster I would try other solution pop up 2 temporary server that have EVC enabled and migrate each host one by one.
If you found this information useful, please consider awarding points for "Correct" or "Helpful". Thanks!!!
Regards,
Stefan Nguyen
VMware vExpert 2009
iGeek Systems Inc.
VMware, Citrix, Microsoft Consultant
Well, its a good standard operation checkpoint whenever you architect new solution and make sure everything is checked before deploying bunch of VMs in a cluster. Always plan and architect accordingly that's the key.
The new hardware wasn't purchased to be a replacement cluster, it was purchased to be a second cluster and it was architected perfectly. Swapping the old hardware for the new was an after thought as a means to refresh the existing hardware while we had the capacity.
If its a big cluster I would try other solution pop up 2 temporary server that have EVC enabled and migrate each host one by one.
I like that idea but It appears that the existing hardware is not EVC compaliant.
________________________________
Jason D. Langdon