Following an upgrade of the vCenter environment from version 9.0.1 to 9.1.0.100, users may encounter the following issues:
vCenter 9.1.0.0100
vSphere Kubernetes Service
There was a Supervisor cluster that had not been cleanly removed and the Supervisor cluster entry remaining in the vCenter database (WCP schema). The Workload Control Plane service (wcpsvc) maintains an in-memory catalog of Supervisor clusters; however, the presence of an orphaned entry causes a lookup failure during catalog initialization.
When this occurs, the wcpsvc.log will contain error messages similar to the following: error fetching edges: ... supervisor '<UUID>' does not exist
Because the service cannot resolve the nonexistent Supervisor ID referenced in the database, it triggers the HTTP 500 errors for API calls and prevents valid Supervisor clusters from being rendered in the vCenter UI.
You can confirm this by running the following command on the vCenter appliance
/usr/lib/vmware-wcp/decryptK8Pwd.py
There will be a cluster entry with a domain-c<ID> that does not correspond to an active cluster in the inventory.
Please open a support case with Broadcom Support to resolve this issue.