How is a headless service created in Kubernetes?

How is a headless service created in Kubernetes?

• A. Setting the service type to NodePort
• B. Deleting the ClusterIP field
• C. Setting the ClusterIP to 'None'
• D. Setting the service type to LoadBalancer

Answer: (C)

Setting the cluster IP to None creates a headless service. When you configure clusterIP as None, Kubernetes does not allocate a virtual IP for the service, so DNS returns the actual pod endpoints instead of a single cluster IP. This enables clients to connect directly to individual pods, which is useful for stateful workloads or custom service discovery. Choosing a NodePort or LoadBalancer does not make a service headless because those types still allocate a cluster IP and expose the service via external endpoints or node ports. Simply removing or omitting the clusterIP field won't guarantee a headless service—the API defaults to assigning a cluster IP unless you explicitly set it to None.

Setting the cluster IP to None creates a headless service. When you configure clusterIP as None, Kubernetes does not allocate a virtual IP for the service, so DNS returns the actual pod endpoints instead of a single cluster IP. This enables clients to connect directly to individual pods, which is useful for stateful workloads or custom service discovery.

Choosing a NodePort or LoadBalancer does not make a service headless because those types still allocate a cluster IP and expose the service via external endpoints or node ports. Simply removing or omitting the clusterIP field won't guarantee a headless service—the API defaults to assigning a cluster IP unless you explicitly set it to None.