vSAN 8.x
The repurposed host retained stale disk groups and metadata from a previous vSAN cluster. The current cluster detects this foreign metadata and flags the referenced, but non-existent objects as inaccessible.
To resolve this issue, both reactive and proactive cleanup steps are performed.
Reactive Cleanup (Optional)
If inaccessible objects are currently blocking cluster operations, the following steps are taken to identify and remove them:
esxcli vsan debug object list --health=inaccessible/usr/lib/vmware/osfs/bin/objtool delete -u [Object_UUID]Note: Exercise caution, as objtool delete is an irreversible action. Ensure the object is confirmed to be stale data from a previous cluster before proceeding. If doubt exists regarding the state of the object, engage Broadcom Support, as premature deletion results in potential data loss.
Proactive Cleanup (Recommended)
To permanently clear stale configuration and prevent recurrent alarms: