Harbor Addon Stuck in "Deleting" State in TCA Manager GUI Despite Backend Removal

Products

• VMware Telco Cloud Automation

Issue/Introduction

The Harbor addon appears stuck in a "Deleting" state in the Telco Cloud Automation (TCA) UI.

TCA Manager logs (app.log or tca_cp logs) show timeout errors connecting to the cluster: ClusterConnectionReady:failed to create cluster accessor... Client.Timeout exceeded while awaiting headers

Validating the backend via kubectl confirms the Harbor resource is already gone.

Environment

3.2

Cause

The issue is caused by a stale entry in the TCA Manager database. A network timeout or connectivity interruption between TCA Manager and the workload cluster prevented the system from processing the final deletion event, leaving the database row orphaned while the actual Kubernetes resource was successfully removed.

Resolution

Prerequisites:

• Ensure the Harbor is actually deleted from the backend by running: kubectl get tka -A in the management cluster (Ensure no Harbor addon is listed for the affected cluster).
• Important: Take a snapshot of the TCA Manager appliance before performing database operations.

Procedure:

1. SSH into the TCA Manager appliance as admin
2. Connect to the Postgres database pod:
   

   kubectl exec -it postgres-0 -n tca-mgr -- psql -d tca -U tca_admin -h localhost

3. Replace <your_cluster_name> with the actual name of the cluster as it appears in TCA.
   

   delete from "K8sClusterDetails" where val->>'rowType'='addOn' and val->'metadata'->>'name'='harbor' and val->'metadata'->>'clusterName'='your_cluster_name';
   

4. Exit the psql session:
5. Refresh the TCA Manager GUI. The Harbor addon should no longer appear.