Configuring VirtualServer with custom HTTP and HTTPS listener ports

This tutorial outlines how to configure and deploy a VirtualServer resource with custom HTTP and HTTPS listener ports.

Configuring a VirtualServer with custom HTTP and HTTPS listener ports.

VirtualServer can explicitly define custom HTTP and HTTPS listener ports using the spec.listener.http and spec.listener.https fields. Each field must reference a valid listener defined by in a GlobalConfiguration resource.

Deploy GlobalConfiguration

Create a yaml file called nginx-configuration.yaml with the below content:

apiVersion: k8s.nginx.org/v1
kind: GlobalConfiguration
metadata:
  name: nginx-configuration
  namespace: nginx-ingress
spec:
  listeners:
  - name: http-8083
    port: 8083
    protocol: HTTP
  - name: https-8443
    port: 8443
    protocol: HTTP
    ssl: true

Deploy nginx-configuration.yaml file:

kubectl apply -f nginx-configuration.yaml

Deploying NGINX Ingress Controller with GlobalConfiguration resource

Using Helm
Using Manifests

Add the below arguments to the values.yaml file in controller.globalConfiguration:

spec:
  listeners:
  - name: http-8083
    port: 8083
    protocol: HTTP
  - name: https-8443
    port: 8443
    protocol: HTTP
    ssl: true

Follow the Installation with Helm instructions to deploy the NGINX Ingress Controller with custom resources enabled.

Ensure your NodePort or LoadBalancer service is configured to expose the custom listener ports. This is set in the customPorts section under controller.service.customPorts:

customPorts:
  - name: http-8083
    port: 8083
    protocol: TCP
    targetPort: 8083
  - name: https-8443
    port: 8443
    protocol: TCP
    targetPort: 8443

Add the below argument to the manifest file of NGINX Ingress Controller:
```
args:
  - -$(POD_NAMESPACE)/nginx-configuration
```
Follow the Installation with Manifests instructions to deploy NGINX Ingress Controller with custom resources enabled.

Ensure your NodePort or LoadBalancer service is configured to expose the custom listener ports. Below is an example yaml configuration using NodePort, which would also apply to a LoadBalancer service:

apiVersion: v1
kind: Service
metadata:
  name: nginx-ingress
  namespace: nginx-ingress
spec:
  type: NodePort
  ports:
  - port: 8083
    targetPort: 8083 # Custom HTTP listener port
    protocol: TCP
    name: http-8083
  - port: 8443
    targetPort: 8443 # Custom HTTPS listener port
    protocol: TCP
    name: https-8443
  selector:
    app: nginx-ingress

Deploying VirtualServer with custom listeners

Deploy the custom listeners resources from the repository examples. It includes all required resources, including VirtualServer.

Below is a snippet of the VirtualServer resource that will be deployed:

apiVersion: k8s.nginx.org/v1
kind: VirtualServer
metadata:
  name: cafe
spec:
  listener:
    http: http-8083
    https: https-8443
  host: cafe.example.com
  tls:
    secret: cafe-secret
  upstreams:
    ...

Below is a snippet of the NGINX configuration for this VirtualServer.

server {
    listen 8083;
    listen [::]:8083;

    server_name cafe.example.com;

    set $resource_type "virtualserver";
    set $resource_name "cafe";
    set $resource_namespace "default";

    listen 8443 ssl;
    listen [::]:8443 ssl;

    ssl_certificate /etc/nginx/secrets/default-cafe-secret;
    ssl_certificate_key /etc/nginx/secrets/default-cafe-secret;
}

Testing custom listener ports

You can test that the VirtualServer resource is deployed with non-default port configuration by explicitly defining them when sending requests.

curl using port 8443:

curl -k https://cafe.example.com:8443/coffee

Server address: 10.32.0.40:8080
Server name: coffee-7dd75bc79b-qmhmv
...
URI: /coffee
...

curl using port 8083:

curl -k http://cafe.example.com:8083/coffee

Server address: 10.32.0.40:8080
Server name: coffee-7dd75bc79b-qmhmv
...
URI: /coffee
...