2025 Updates Introduction to Gateway API

In this lesson, we introduce the Gateway API, a modern, more flexible approach to managing network routing in Kubernetes. In previous discussions on Ingress, we saw a scenario where two services shared the same Ingress resource:

The image illustrates a network architecture diagram showing ingress limitations, with separate paths for "wear-service" and "video-service" leading to multiple instances of "wear" and "video."

Imagine a scenario where independent teams or even separate organizations manage different services. For example, one team could be responsible for the web service while another oversees the video service. In such multi-tenant environments, a single Ingress resource — which can be controlled by only one team at a time — may lead to coordination challenges and potential conflicts. In addition, Ingress has limited support for multi-tenancy and parameterized rules.

For basic routing, consider this simple Ingress configuration:

apiVersion: extensions/v1beta1
kind: Ingress
metadata:
  name: ingress-wear-watch
spec:
  rules:
    - host: wear.my-online-store.com
      http:
        paths:
          - backend:
              serviceName: wear-service
              servicePort: 80
    - host: watch.my-online-store.com
      http:
        paths:
          - backend:
              serviceName: watch-service
              servicePort: 80

Another limitation of Ingress is its narrow rule configuration. Ingress only supports HTTP-based requests by matching hosts or paths. It does not natively support protocols such as TCP or UDP, nor does it offer advanced features like traffic splitting, header manipulation, authentication, or rate limiting. These behaviors are usually implemented using controller-specific annotations. For example:

apiVersion: extensions/v1beta1
kind: Ingress
metadata:
  name: ingress-wear-watch
  annotations:
    wear.my-online-store.com:
      nginx.ingress.kubernetes.io/ssl-redirect: "true"
      nginx.ingress.kubernetes.io/force-ssl-redirect: "true"
spec:
  rules:
    - host: wear.my-online-store.com
      http:
        paths:
          - path: /foo
            backend:
              serviceName: wear-service
              servicePort: 80
    - host: watch.my-online-store.com
      http:
        paths:
          - backend:
              serviceName: watch-service
              servicePort: 80

The use of annotations, however, creates challenges. Different Ingress implementations (such as NGINX and Traefik) require their own annotation syntax. For instance, configuring CORS might look like this:

apiVersion: networking.k8s.io/v1
kind: Ingress
metadata:
  name: cors-ingress
  annotations:
    nginx.ingress.kubernetes.io/enable-cors: "true"
    nginx.ingress.kubernetes.io/cors-allow-methods: "GET, PUT, POST"
    nginx.ingress.kubernetes.io/cors-allow-origin: "https://allowed-origin.com"
    nginx.ingress.kubernetes.io/cors-allow-credentials: "true"

apiVersion: networking.k8s.io/v1
kind: Ingress
metadata:
  name: traefik-ingress
  annotations:
    # CORS Configuration
    traefik.ingress.kubernetes.io/headers.customresponseheaders: |
      Access-Control-Allow-Origin: '*'
      Access-Control-Allow-Methods: GET,POST,PUT,DELETE,OPTIONS
      Access-Control-Allow-Headers: Content-Type,Authorization
      Access-Control-Allow-Credentials: true
      Access-Control-Max-Age: 3600

Kubernetes passes these annotations to the controllers without validating them, meaning that similar configurations will only work with the controllers that recognize the given annotations. This is where Gateway API steps in. Gateway API is an official Kubernetes project designed to improve both layer 4 and layer 7 routing. It represents the next evolution in Kubernetes Ingress, load balancing, and service mesh APIs.

Benefits of Gateway API

One of the key limitations with Ingress is multi-tenancy. Gateway API addresses this by splitting the responsibilities among three objects, each managed by a different persona:

Infrastructure Providers: They define the GatewayClass, which specifies the underlying network infrastructure (e.g., NGINX, Traefik, or another load balancer).
Cluster Operators: They create Gateway objects as instances of a GatewayClass.
Application Developers: They configure HTTPRoutes (or other routes like TCPRoute or gRPCRoute) that attach to the Gateway.

Unlike Ingress, which only supports HTTP routing, Gateway API broadens support to include multiple protocols. Let’s look at how the components are defined.

Defining the GatewayClass

The GatewayClass object specifies the network infrastructure controller:

apiVersion: gateway.networking.k8s.io/v1
kind: GatewayClass
metadata:
  name: example-class
spec:
  controllerName: example.com/gateway-controller

Creating a Gateway

Next, define a Gateway that references the GatewayClass and sets up an HTTP listener on port 80:

apiVersion: gateway.networking.k8s.io/v1
kind: Gateway
metadata:
  name: example-gateway
spec:
  gatewayClassName: example-class
  listeners:
    - name: http
      protocol: HTTP
      port: 80

Configuring an HTTPRoute

Application developers then create an HTTPRoute to define routing rules. In the following example, requests for “www.example.com” with a path prefix of “/login” are directed to the “example-svc” backend on port 8080:

apiVersion: gateway.networking.k8s.io/v1
kind: HTTPRoute
metadata:
  name: example-httproute
spec:
  parentRefs:
    - name: example-gateway
  hostnames:
    - "www.example.com"
  rules:
    - matches:
        - path:
            type: PathPrefix
            value: /login
      backendRefs:
        - name: example-svc
          port: 8080

This declarative configuration explicitly defines both matching rules and backend mapping, eliminating the reliance on ambiguous annotations. Additionally, it paves the way for supporting protocols beyond HTTP, such as TCP or gRPC.

Streamlined TLS Termination

Another advantage of Gateway API is its clear and structured approach to TLS configuration. With Ingress, configuring HTTPS typically involves both native TLS settings and controller-specific annotations:

# Ingress configuration for HTTPS using annotations
apiVersion: networking.k8s.io/v1
kind: Ingress
metadata:
  name: secure-app
  annotations:
    nginx.ingress.kubernetes.io/ssl-redirect: "true"
    nginx.ingress.kubernetes.io/force-ssl-redirect: "true"
spec:
  tls:
  - hosts:
      - secure.example.com
    secretName: tls-secret

The equivalent configuration in Gateway API is both straightforward and self-contained:

apiVersion: gateway.networking.k8s.io/v1
kind: Gateway
metadata:
  name: secure-gateway
spec:
  gatewayClassName: example-gc
  listeners:
    - name: https
      port: 443
      protocol: HTTPS
      tls:
        mode: Terminate
        certificateRefs:
          - kind: Secret
            name: tls-secret
      allowedRoutes:
        kinds:
          - kind: HTTPRoute

Here, TLS termination is explicitly defined with certificate references and allowed routes for the listener, eliminating the need for external annotations.

Traffic Splitting for Canary Deployments

Ingress can handle traffic splitting through controller-specific annotations. For example, a canary deployment might be configured as follows:

# Canary deployment configuration with Ingress annotations for NGINX
apiVersion: networking.k8s.io/v1
kind: Ingress
metadata:
  name: canary-ingress
  annotations:
    nginx.ingress.kubernetes.io/canary: "true"
    nginx.ingress.kubernetes.io/canary-weight: "20"
spec:
  rules:
    - http:
        paths:
          - path: /
            pathType: Prefix
            backend:
              service:
                name: app-v2
                port:
                  number: 80

With Gateway API, traffic splitting is declared directly in the HTTPRoute, making it more transparent:

apiVersion: gateway.networking.k8s.io/v1
kind: HTTPRoute
metadata:
  name: split-traffic
spec:
  parentRefs:
    - name: app-gateway
  rules:
    - backendRefs:
        - name: app-v1
          port: 80
          weight: 80
        - name: app-v2
          port: 80
          weight: 20

This clear definition of traffic distribution simplifies canary deployments without relying on hidden annotations.

Advanced Configurations: CORS

Managing advanced configurations such as CORS with Ingress often requires multiple annotations. For example, for NGINX:

# CORS configuration for NGINX using annotations
apiVersion: networking.k8s.io/v1
kind: Ingress
metadata:
  name: cors-ingress
  annotations:
    nginx.ingress.kubernetes.io/enable-cors: "true"
    nginx.ingress.kubernetes.io/cors-allow-methods: "GET, PUT, POST"
    nginx.ingress.kubernetes.io/cors-allow-origin: "https://allowed-origin.com"
    nginx.ingress.kubernetes.io/cors-allow-credentials: "true"

Or for Traefik:

apiVersion: networking.k8s.io/v1
kind: Ingress
metadata:
  name: traefik-ingress
  annotations:
    # CORS Configuration
    traefik.ingress.kubernetes.io/headers.customresponseheaders: |
      Access-Control-Allow-Origin: '*'
      Access-Control-Allow-Methods: GET, POST, PUT, DELETE, OPTIONS
      Access-Control-Allow-Headers: Content-Type,Authorization
      Access-Control-Allow-Credentials: true
      Access-Control-Max-Age: 3600

With Gateway API, CORS is configured in a centralized, declarative manner using filters:

apiVersion: gateway.networking.k8s.io/v1
kind: HTTPRoute
metadata:
  name: cors-route
spec:
  parentRefs:
    - name: my-gateway
  rules:
    - matches:
        - path:
            type: PathPrefix
            value: /api
      filters:
        - type: ResponseHeaderModifier
          responseHeaderModifier:
            add:
              - name: Access-Control-Allow-Origin
                value: "*"
              - name: Access-Control-Allow-Methods
                value: "GET, POST, PUT, DELETE, OPTIONS"
              - name: Access-Control-Allow-Headers
                value: "Content-Type,Authorization"
              - name: Access-Control-Allow-Credentials
                value: "true"
              - name: Access-Control-Max-Age
                value: "3600"
      backendRefs:
        - name: api-service

This configuration works consistently across Gateway API implementations regardless of the underlying controller, ensuring that advanced settings like CORS remain centralized and transparent.

Controller Support and Final Remarks

Today, many controllers have already implemented (or are in the process of implementing) Gateway API. Notable examples include Amazon EKS, Azure Application Gateway for Containers, Contour, Envoy, Google Kubernetes Engine, HAProxy, Istio, Kong, Kuma, NGINX, and others. That concludes this lesson on the Gateway API. You can now practice working with Gateway API controllers in the accompanying labs. Enjoy the enhanced flexibility and clarity that Gateway API brings to your Kubernetes networking configurations!

Introduction

Core Concepts

Scheduling

Logging & Monitoring

Application Lifecycle Management

Cluster Maintenance

Security

Storage

Networking

Troubleshooting

Design and Install a Kubernetes Cluster

Helm Basics (2025 Updates)

Kustomize Basics (2025 Updates)

Mock Exams

Install "Kubernetes the kubeadm way"

2025 Updates Introduction to Gateway API

Benefits of Gateway API

Defining the GatewayClass

Creating a Gateway

Configuring an HTTPRoute

Streamlined TLS Termination

Traffic Splitting for Canary Deployments

Advanced Configurations: CORS

Controller Support and Final Remarks

Watch Video

Practice Lab

​Benefits of Gateway API

​Defining the GatewayClass

​Creating a Gateway

​Configuring an HTTPRoute

​Streamlined TLS Termination

​Traffic Splitting for Canary Deployments

​Advanced Configurations: CORS

​Controller Support and Final Remarks

Watch Video

Practice Lab

Benefits of Gateway API

Defining the GatewayClass

Creating a Gateway

Configuring an HTTPRoute

Streamlined TLS Termination

Traffic Splitting for Canary Deployments

Advanced Configurations: CORS

Controller Support and Final Remarks