We have a host with 16 CPUs with 2 VMs hosted on it (each VM has 8 CPUs). From time to time we have experience 100% CPU on one of the VMs yes the load on the host is normal.
vmotioning the VMs to different hosts doesnt help.
We've update bio/firmware & drivers for the host and ran hardware diag to confirm the hosts are OK.
Is there anything else we can check?