A vSphere HA-enabled cluster is configured with a host isolation response of "Shut Down and restart VMs". A host in this cluster has a single NIC on the management network. If the management NIC on th

When the sole management NIC fails, the host loses communication with the rest of the HA cluster and is declared isolated. The 'Shut Down and restart VMs' response invokes VMware Tools on each running VM to perform a graceful OS shutdown - distinct from a hard power-off - and then HA restarts those VMs on other hosts that are still active in the cluster.

vMotion requires an active network connection between source and destination hosts to transfer VM memory and state; a host with a failed management NIC cannot coordinate or complete a vMotion operation.

An immediate power-off without a graceful shutdown describes the 'Power Off and restart VMs' isolation response, not 'Shut Down and restart VMs', which specifically uses VMware Tools for a clean guest shutdown.

With a single management NIC failure the host is fully isolated from the cluster heartbeat network, which triggers the configured HA isolation response rather than allowing VMs to keep running on the isolated host.