vSAN Skyline Health displays an alert for .vsan.stats (Missing).
Investigation of the vSAN datastore reveals duplicate .vsan.stats objects.
One object is actively updating and in a healthy state, while the duplicate object is in an Inaccessible (APD) state.
VMware vSAN 7.x
VMware vSAN 8.x
VMware vSAN 9.x
This issue is caused by an orphaned performance statistics object remaining in the vSAN datastore namespace (CMMDS).
This typically occurs following a cluster restart or host failure where vSAN provisions a new statistics object but fails to clear the stale entry.
The Skyline Health service detects the orphaned APD object, triggering the missing stats alert despite the presence of a functioning, healthy object.
Log in to any ESXi host in the affected vSAN cluster via SSH using the root account.
Identify the Object UUID of the inaccessible .vsan.stats object by executing: esxcli vsan debug object list --health=inaccessible
Locate the .vsan.stats namespace in the output and copy the associated Object UUID. Verify that the UUID belongs to the inaccessible object and not the healthy production object.
Forcefully delete the orphaned object using the objtool utility (replace <Object-UUID> with the UUID identified ): /usr/lib/vmware/osfs/bin/objtool delete -u <Object-UUID> -f
Log in to the vSphere Client.
Navigate to the cluster > Monitor > vSAN > Skyline Health.
Click Retest to refresh the health checks and verify the alert is resolved.
For scenarios where no healthy stats object exists and the service must be recreated, refer to KB 387843 (Stats Primary election alert Triggered after vSAN cluster restart). For general inaccessible object troubleshooting, refer to KB 389599.