Loading auth...
Medium10 minContainer Orchestration
UpdatedAug 2, 2026
Edit

Kubernetes Service Types

CONCEPTS:Kubernetes Architecture

Why This Is Asked

Exposing applications is a core task in Kubernetes. This question tests whether a candidate understands the different ways to route traffic to Pods and when to use internal vs. external exposure.

Key Concepts

  • ClusterIP: Internal-only communication.
  • NodePort: Exposing a service on a static port on each Node’s IP.
  • LoadBalancer: Using a cloud provider’s external load balancer.
  • ExternalName: Mapping a service to a DNS name.

Answers by Technology

+ Add Variant
KubernetesImprove this answer ✏️

Expected Answer

Kubernetes Services provide a stable network identity (IP and DNS name) for a set of Pods. The four main types are:

  1. ClusterIP (Default): Exposes the Service on a cluster-internal IP. Use this for internal communication between microservices.
  2. NodePort: Exposes the Service on each Node’s IP at a static port (usually 30000-32767). You can reach the service from outside by calling <NodeIP>:<NodePort>.
  3. LoadBalancer: Exposes the Service externally using a cloud provider’s load balancer (e.g., AWS ELB, GCP LB). It automatically creates a NodePort and ClusterIP for you.
  4. ExternalName: Maps the Service to the contents of the externalName field (e.g., my.database.example.com), by returning a CNAME record.

Why It Matters

Using the wrong service type can lead to security vulnerabilities (unintentionally exposing a private DB) or architectural complexity. Most production environments use ClusterIP for internal traffic and an Ingress Controller (which sits on top of a LoadBalancer service) to manage external traffic.

Example Code

A LoadBalancer Service

apiVersion: v1
kind: Service
metadata:
  name: my-web-service
spec:
  type: LoadBalancer
  selector:
    app: my-web-app
  ports:
    - protocol: TCP
      port: 80
      targetPort: 8080

Common Mistakes

  • Using NodePort for production: It’s hard to manage because you need to track Node IPs and it opens up a wide range of ports on your servers.
  • Forgetting the Selector: If the selector doesn’t match the Pod labels, the Service will have no endpoints and will return 404/Timeout.
  • Not using Ingress for multiple services: Creating a separate LoadBalancer service for every single microservice is expensive and hard to manage (use one Ingress instead).

Follow-up Questions

  • What is an Ingress? (Answer: An API object that manages external access to the services in a cluster, typically HTTP. It provides load balancing, SSL termination, and name-based virtual hosting).
  • What is a Headless Service? (Answer: A Service with clusterIP: None. It doesn’t provide a single IP, but instead returns the A records (IPs) for all the backing Pods. Used for stateful applications like databases).
  • What is a TargetPort? (Answer: The port on which the application inside the container is listening).

References