An incident is resolved by restarting a service, but the same failure recurs every week. What process should address the underlying cause?

1. Only a change to the service's display name
2. A decision to stop recording incidents
3. Problem management
4. Only incident closure administration