Certified Kubernetes Application Developer (CKAD) Practice Test

The feature which enables Kubernetes to recover from failed pods automatically is known as self-healing capabilities. This functionality is intrinsic to Kubernetes and is a significant component of its orchestration abilities. When a pod fails or goes down, Kubernetes continuously monitors the state of the pods using a component called the kubelet. If it detects that a pod is not functioning correctly, it can automatically restart the pod or replace it with a new instance, ensuring high availability and resilience for the applications running within the cluster.

Self-healing works through the use of health checks, such as readiness and liveness probes, which allow Kubernetes to determine the health of the pods at runtime. This means that the system can automatically respond to issues without requiring any manual intervention, significantly improving the reliability of applications deployed in a Kubernetes environment.

While other options like load balancing, horizontal scaling, and access control are important aspects of Kubernetes, they do not specifically address the ability to recover from pod failures. Load balancing distributes traffic among available pods, horizontal scaling adjusts the number of pod replicas based on demand, and access control manages permissions within the cluster. However, none of these features inherently provide the automatic recovery ability described in the question, making self-healing the correct and relevant choice.