Kyma

Appearance

You Cannot Delete Leftover Service Instances and Bindings ​

Symptom ​

After your Kyma cluster has become unavailable, some orphaned service instances and service bindings are still present in the SAP Business Technology Platform (BTP).

Cause ​

You cannot access your Kyma cluster where the service instances and bindings were created. Without access to the cluster, you cannot use standard deletion methods.

Solution ​

WARNING

This cleanup procedure deletes all service instances and bindings associated with the inaccessible cluster.

  1. To access the Service Manager API, perform the following actions:

    1. In the SAP BTP cockpit, create an SAP Service Manager service instance with the subaccount-admin plan and its service binding. The service-operator-access plan does not have the necessary access level for performing the cleanup operation. See Creating Instances in Other Environments and Creating Service Bindings in Other Environments.

    2. To get an access token, run the following command. Replace the placeholders with the url, clientid, and clientsecret from the service binding to the Service Manager instance you have created.

      bash
      curl '{url}/oauth/token' -X POST \
   -H 'Accept: application/json' \
   -d 'grant_type=client_credentials&client_id={clientid}&client_secret={clientsecret}'

  2. To prepare the DELETE request, identify the following parameters:

    • platformID - the ID of the Service Manager instance with the service-operator-access plan.
    • clusterID - the ID of your cluster. To get clusterID, go to the Instances section in the SAP BTP cockpit, and copy it from the Scope column in the row that provides information about the service instance(s) you want to delete.

  3. Send the following request:

    bash
    curl -X DELETE '{sm_url}v1/platforms/{platformID}/clusters/{clusterID}' \
   -H 'Authorization: Bearer {ACCESS_TOKEN}'

  4. Monitor the response. You can get one of the following options:

    • 202 Accepted - the request is accepted for processing.
    • 404 Resource Not Found - platform or cluster not found.
    • 429 Too Many Requests - the rate limit exceeded.

  5. For the operation status, review the following headers:

    • Location - the path provided to monitor the status of the operation. For more information about operations, see Service Manager operation API.
    • Retry_After - indicates the time in seconds after which you can retry the request after hitting rate limits.