Hey community!
Anyone running 7.0u1 successfully in a nested lab?
I have a nested cluster running ontop of a (physical) 6.7u3 ESXi host.
The vCLS VM is created but fails to power on with this task error:
"Feature 'MWAIT' was absent, but must be present"
And the event log shows:
"Cluster Agent VM cannot be powered on due to insufficient resources on cluster"
Deleting the VM (which forces a recreate) or even a new vSphere cluster creation always ends with the same problem.
The hw ESXi hosts underneath has an quite new Intel Xeon Gold CPU and the nested ESX VMs have the "Expose hardware assisted virtualization to the guest OS" box checked in the VM settings.
Any hint how to get this running?
Thx!!
Mario
Ok figured this out on my own after some testing.
The ESXi box underneath is a Lenovo server, which had (I guess by default) MONITOR/MWAIT flag in the bios disabled.
After enabling it, it seems to be also exposed to the guest OS. And so the vCLI VMs on the nested ESXi hosts are now up and running.
Ok figured this out on my own after some testing.
The ESXi box underneath is a Lenovo server, which had (I guess by default) MONITOR/MWAIT flag in the bios disabled.
After enabling it, it seems to be also exposed to the guest OS. And so the vCLI VMs on the nested ESXi hosts are now up and running.
Nice find. Monitor/Mwait is indeed a requirement for EVC, and EVC is enabled on a per VM basis for vCLS.
is there a VMWRAE KB document refer to this issue?