What Can I Do If an MCS or MCI Instance Fails to Be Deleted?

Symptom: The MCS or MCI instance is always in the Deleting state.

Solution: Forcibly delete the abnormal instances and manually clear residual resources.

Procedure

1. Log in to the UCS console. In the navigation pane, choose Fleets. On the Fleets tab, click the name of the target fleet to access its details page.
2. In the navigation pane, choose Services and Ingresses. On the displayed page, click the MCS or MCI tab.
3. Select the abnormal instance that is always in the Deleting state.
4. Manually clear the residual ELB resources of this abnormal instance.
   1. Click under the access type of the abnormal instance. The details page of the load balancer associated with this instance is displayed.
   2. On the Summary tab, disable Deletion Protection.

      

   3. Switch to the Listeners tab and delete the listener associated with the abnormal instance.
   4. In the navigation pane, choose Backend Server Groups. On the displayed page, click the name of the backend server group associated with the listener of the abnormal instance. On the Summary tab, disable Removal Protection and Modification Protection.

      

   5. Delete the backend server group associated with the listener of the abnormal instance.

5. Forcibly delete the abnormal instance.
   1. Select the abnormal instance that is always in the Deleting state and click Edit YAML in the Operation column.
   2. Delete the following content from the YAML file to forcibly delete the instance:

      finalizers:

      - karmada.io/multi-cluster-service-controller ## MCS

      finalizers:

      - karmada.io/multi-cluster-ingress-controller ## MCI