Which Kubernetes feature helps maintain desired state by restarting failed pods?

• A. Rolling updates
• B. Service discovery
• C. Self-healing
• D. Ecosystem integration

Answer: (C)

Self-healing is the mechanism that keeps applications running by automatically restarting or recreating pods that fail so the declared number of replicas is maintained. In Kubernetes, you describe the desired state (for example, a Deployment that specifies how many replicas should run). The control plane watches the actual state, and if a pod crashes or becomes unhealthy (detected through liveness probes or mere failure), the appropriate controller automatically starts a new pod or restarts the failed one to bring the system back to the desired state. This automatic recovery is what keeps services available without manual intervention. Rolling updates focus on updating versions, service discovery helps clients find pods, and ecosystem integration refers to broader toolchain connections—they don’t directly ensure pods are restarted to maintain the desired state.