Which feature of Deployments ensures the number of Pods is as expected in case of a Pod crash or deletion?

Which feature of Deployments ensures the number of Pods is as expected in case of a Pod crash or deletion?

• A. Scheduling
• B. Replication
• C. Networking
• D. Scaling

Answer: (B)

When you set a Deployment, you declare a desired number of replicas. Kubernetes continuously reconciles the actual state with that desired state. If a Pod crashes or is deleted, the Deployment’s ReplicaSet notices there are fewer Pods than requested and creates new Pods to bring the count back up. This automatic replacement is the replication mechanism at work, keeping the number of Pods steady even after failures. Scheduling would decide which node runs a Pod, not how many Pods should exist. Networking isn’t about Pod counts. Scaling refers to changing the desired number, but the automatic maintenance of the exact replica count after a crash is provided by replication through the Deployment and its ReplicaSet.

When you set a Deployment, you declare a desired number of replicas. Kubernetes continuously reconciles the actual state with that desired state. If a Pod crashes or is deleted, the Deployment’s ReplicaSet notices there are fewer Pods than requested and creates new Pods to bring the count back up. This automatic replacement is the replication mechanism at work, keeping the number of Pods steady even after failures.

Scheduling would decide which node runs a Pod, not how many Pods should exist. Networking isn’t about Pod counts. Scaling refers to changing the desired number, but the automatic maintenance of the exact replica count after a crash is provided by replication through the Deployment and its ReplicaSet.