In OpenShift 4 an IngressController is degraded because the LoadBalancer service is pending

Issue

The cluster upgrade is halted because of the error:

$ oc -n openshift-ingress-operator logs ingress-operator-7b4fdfc769-pvfdg -c ingress-operator
...
2024-10-04T18:33:24.206986862Z 2024-10-04T18:33:24.206Z ERROR   operator.ingress_controller controller/controller.go:119    got retryable error; requeueing {"after": "1m0s", "error": "IngressController is degraded: LoadBalancerReady=False (LoadBalancerPending: The LoadBalancer service is pending)"}