Prerequisites
- Please take simultaneous non-memory snapshots of each virtual appliance(s) in the cluster.
- Use the quiesce option for versions 8.9.x and above.
- You have SSH access with root user and password.
Procedure
- SSH to the appliance using the root user account.
- Run the following command to connect to the vco-db
vracli dev psql vco-db
- Verify the workflow(s) that has the 'v' flag by running this query:
SELECT id,name FROM vmo_workflow WHERE allowedoperations = 'v';
- Update the workflow(s) that have only the 'v' flag to also have the 'e' flag
UPDATE vmo_workflow SET allowedoperations = 've' WHERE allowedoperations = 'v';
- Exit the Postgres database:
\q
- Restart all the orchestrator pods:
kubectl -n prelude delete pod -l app=vco-app
- Monitor the restart progress until you have all the pods running:
kubectl -n prelude get pods -l app=vco-app -w
- Press Ctrl + C to cancel the watch command from Step #8.
Example:
- Delete the workflow(s) that are experiencing the problem.