When attempting an ESXi upgrade, you may encounter a failure due to "ESXi cluster is not healthy" errors, despite all prechecks passing during the planning phase. This issue typically occurs when there are active "vSphere HA failover" alerts in vCenter.
Error seen:
“The ESXi cluster is not healthy (green) due to : vSphere HA failover operation in progress in cluster <Cluster Name> in datacenter <Datacenter Name>: <VMCount> VMs being restarted, <VMCount> VMs waiting for a retry, <VMCount> VMs waiting for resources, <VMCount> inaccessible vSAN VMs
VCF 4.x
VCF 5.x
These alerts often occur when a host fails, and vSphere HA attempts to restart a VM that was already powered off.
To resolve this issue: