In previous chapters, you learned to run applications using Pods and manage their lifecycle with Deployments. However, Pods are ephemeral and their IP addresses are not stable, which presents a challenge for communication. How do your application's components find and talk to each other reliably? How do you expose your application to users outside the cluster? This chapter focuses on the Kubernetes networking model designed to solve these problems.
You will start by examining the principles of communication within a cluster. We will then introduce the Service object, an abstraction that provides a durable network endpoint for a set of Pods. You will learn to differentiate and use various Service types, including ClusterIP, NodePort, and LoadBalancer, to control how your application is exposed. Finally, we will cover Ingress, a resource that manages external HTTP and HTTPS access to services in the cluster.
4.1 The Kubernetes Networking Model
4.2 Introduction to Services
4.3 Service Types: ClusterIP, NodePort, and LoadBalancer
4.4 Defining a Service Manifest
4.5 Service Discovery within the Cluster
4.6 Exposing Services with Ingress
4.7 Hands-on Practical: Exposing an Application
© 2026 ApX Machine LearningEngineered with