How to Set Up Automatic Failover with VMware High Availability

To configure automatic failover, first create a cluster in the vSphere Client and enable vSphere HA during cluster setup. Then add at least two ESXi hosts to the cluster. The hosts must share storage and have correctly configured management networking so they can exchange heartbeats.

After adding the hosts, configure the HA settings for the cluster. The main options are:

Host Failure Response: whether VMs are automatically restarted on another host after a host failure.

Host Isolation Response: how VMs behave if a host loses network connectivity but stays powered on.

Datastore Failure Handling (PDL/APD): how VMs behave during storage outages.

VM Monitoring: restarts VMs automatically if the guest OS stops responding.

Admission Control: keeps enough resources reserved in the cluster to restart VMs after a failure.

Heartbeat Datastores: adds another way to confirm host status.

Once configured, HA keeps monitoring cluster health. If a host crashes or becomes unreachable, HA restarts its VMs on other available hosts, which keeps downtime short and services running.

HA protects against host-level failures, but it does not replace backup. Use failover together with regular VM backups to cover both availability and data protection.

Q1: What happens when a host fails?

HA detects the failure and automatically restarts the affected VMs on another host in the cluster.

Q2: What is the difference between HA and Fault Tolerance?

HA restarts VMs after a failure, while Fault Tolerance runs a live secondary VM to avoid downtime.

Q3: How many hosts are required for HA?

You need at least two hosts to enable failover.

Q4: Does HA protect against data corruption?

No. HA handles availability. It does not provide backup or long-term data protection.