VCSA 6.5 - Power on failure after ESXi host upgrade

[danstr1]

My environment is small with only two hosts running on a shared DAS. Recently, I successfully migrated by Windows-based vCenter 5.5 to VCSA 6.5. I then proceeded by upgrading my hosts to ESXi 6.5. Both hosts upgraded without issue. After upgrdading the host to 6.5 where my VCSA is hosted, I'm now unable to power the vCenter appliance on. The error message below shortly after I opt to power on. My environment remains functional but I'm without a vCenter server until I figure out a workaround. I've created a support case on the matter but I know it may be a while before I hear back. I'm considering installing a new Windows based vCenter server and then connecting my hosts to it. I realize I would need to setup the HA etc again but given the size of my environment, it's a consideration. Any advice is great appreciated! I did see some related KB's and web posts detailing the steps to change the cpuid requirements through vCenter. I don't believe it's an option for me since I can only connect directly. Thanks in advance!

haTask-38-vim.VirtualMachine.powerOn-188336985

Description  

Power On this virtual machine

     Virtual machine:    

LLIVCenter     

State  

  Failed  - The virtual machine feature mask incorrectly requires 'cpuid.STIBP' = '1'.  

Errors  

0">  
• The virtual machine feature mask incorrectly requires 'cpuid.STIBP' = '1'.
• The virtual machine feature mask incorrectly requires 'cpuid.IBPB' = '1'.
• The virtual machine feature mask incorrectly requires 'cpuid.IBRS' = '1'.
• Module 'FeatureCompatLate' power on failed.
• Failed to start the virtual machine

[SupreetK]

Can you share the ESXi build number?

Also, take a backup of the .vmx file and modify it to add the below lines -

1) When in the VM folder, run the command <cp VM_name.vmx VM_name.vmx.backup> to backup the .vmx file.

2) Append the below parameters to the .vmx, either by editing it using vi editor or via the VM advanced configuration parameters option from the host client GUI.

featMask.vm.cpuid.stibp = "Max:0"

featMask.vm.cpuid.ibrs = "Max:0"

featMask.vm.cpuid.ibpb = "Max:0"

3) Reload the VM as shown in VMware Knowledge Base​

4) Try to power on the VM.

If the above step does not work, revert to the previous .vmx before doing any further changes -

1) When in the VM folder, run the command <cp VM_name.vmx VM_name.vmx.changed> followed by <cp VM_name.vmx.backup VM_name.vmx>.

2) Reload the VM.

Please consider marking this answer as "correct" or "helpful" if you think your questions have been answered.

Cheers,

Supreet

[danstr1]

Thanks so much for your response and sorry that I didn't have the opportunity to reply sooner. I spoke with VMWare support, who was able to help me, shortly after posting the question. We connected to the host in question and he ran a few commands against it (wish I had more detail to provide but i don't know the specifics). We then disconnected the VCSA and reconnected it to an alternate host by browsing the datastore and registering it. The VCSA then powered on booted without issue and also migrated between the hosts as well. I'm not sure how much it contributed but it's worth noting, I had also created a new cluster with an EVC policy where the hosts and consequently the VM's were moved.

[SupreetK]

Thanks for the update! Please close the thread accordingly

Cheers,

Supreet