In a VMware vSphere environment, High Availability (HA) is your primary safety net. However, when you see the alert “The number of vSphere HA heartbeat datastores for this host is 0 which is less than required 2,” your safety net has a hole in it.
Key Concept: Heartbeat datastores allow vCenter to determine if a host is truly down or just suffering from a management network isolation. If networking fails, HA uses these datastores to communicate.
Why is there a Datastore Mismatch?
If one host in your cluster is throwing this error while others are not, the discovery process has failed specifically for that host. Common reasons include:
- Zoning Errors: The LUN was not added to the host’s WWN in the FC switch.
- Host Isolation: The host cannot reach the storage network (iSCSI/NFS).
- Inconsistent Presentation: Datastores were mapped to a subset of the cluster instead of all hosts.
Step 1: Discovering Storage via CLI
To identify if the host sees the storage devices at all, run a rescan and check the filesystem list. This bypasses any potential vCenter UI lag.
# Rescan all storage adapters
esxcli storage core adapter rescan --all
# List all mounted filesystems
esxcli storage filesystem list
Step 2: Checking the “Path” Health
If the datastore is missing, check if the underlying paths are active. If they are listed as Dead or Inaccessible, you have a physical layer or zoning problem.
esxcli storage core path list
Step 3: Reconfiguring HA
Once you have restored storage connectivity and the host sees the shared datastores (NFS or VMFS), you must re-initialize the HA agent to recognize the new heartbeat paths:
- Right-click the ESXi Host.
- Select Reconfigure for vSphere HA.
After a few moments, the “0 heartbeat datastores” warning should disappear as the agent selects two shared volumes for heartbeating.
