The easiest way out is to prevent the pod from shutting immediately. This is possible with a preStop hook. Adding in a preStop hook to the deployment which does sleep 5 should delay the pod termination long enough to let the Ingress update and remove the disabled pod from its upstream list.