When a readiness Probe failed, kubernetes remove the Pod from the service point so that traffic could not move toward that pod. Pod remain in the node. Kubernetes continues to run the readiness probe checks against the container. If the application recovers and the readiness probe passes again, Kubernetes automatically re-adds the pod to the service endpoints to resume traffic.
If the pod remain failed in readiness probe, it is of no use in and only consuming the Node CPU and memory. Then why kubernetes do not kill the pod and remove it from the Node? Why there is no time frame to check the readiness probe failure after which the Pod should be removed from the node?