Understand the policy

Review the cluster's selected Admission Control policy and the failover capacity it is designed to reserve. The policy should match the cluster's actual availability objective.

Check current resource demand

Compare CPU and memory demand with host capacity. VM reservations can reduce usable failover capacity even when aggregate cluster utilization appears acceptable.

Check host availability

Confirm that all expected hosts are connected, have compatible configuration and are eligible to run the protected VMs.

Review stale alerts separately

Some HA alerts can remain active because the cluster's calculated failover state has not refreshed. If the cluster demonstrably has sufficient resources, refresh the admission-control calculation rather than assuming an actual capacity shortage.

Check reservations and large VMs

Large reservations and VMs with significant memory requirements can make admission-control calculations fail unexpectedly. Review the largest reservations first.

Validate after changes

After changing capacity or policy, verify that the HA health state updates and that a test or maintenance scenario still satisfies the recovery objective.

Useful commands

Get-Cluster
Get-VMHost
Get-VM
Production workflow: capture the current state, test the dependency directly, make one controlled change, retest the original symptom and document the evidence.

Quick checklist

  • Understand the policy
  • Check current resource demand
  • Check host availability
  • Review stale alerts separately
  • Check reservations and large VMs
  • Validate after changes

Primary reference

This TechRunbook guide is independently written and cross-checked against Broadcom VMware documentation. View the reference →

More VMware troubleshooting

Explore practical VMware, Windows Server, Hyper-V, Azure, PowerShell and MABS runbooks.

Browse all articles →