Install on Amazon EKS

This guide shows you how to install HAProxy Kubernetes Ingress Controller in Amazon Elastic Kubernetes Service. We will use the Helm package manager.

Pre-installation checklist Jump to heading #

• A running EKS Kubernetes cluster with a configured node group as described on the Kubernetes web site.
• Install the AWS CLI from Amazon so you can manage AWS services from your laptop.
• Install the helm command-line tool from the helm web site.
• Install the kubectl command-line tool from the Kubernetes web site.

Connect to your EKS cluster Jump to heading #

1. If you created your EKS cluster using the AWS Console, you were logged into the AWS Console as a certain user. Unless you’ve already configured additional RBAC users inside of your Kubernetes cluster, you must connect to the cluster using this same user. Follow these steps to connect to the cluster as that user:

   • In the AWS Console, expand your account menu, located in the upper right, and select My Security Credentials.
   • On the Your Security Credentials page, expand the Access keys section and create a new access key if you do not already have one stored. This will give you an Access Key ID and Secret Access key. Store these somewhere so that you have them later.

2. On your local workstation, configure a named profile for the AWS CLI to use when connecting to AWS. You can accomplish this by calling the aws configure command, which will prompt you for the necessary values, including your user’s Access Key ID and Secret Access key:

   nix
   aws configure

   nix
   aws configure

3. Connect to your Kubernetes cluster using the AWS CLI. This will create a ~/.kube/config file:

   nix
   aws eks --region [Your region] update-kubeconfig --name [Your cluster name]

   nix
   aws eks --region [Your region] update-kubeconfig --name [Your cluster name]

4. Check that you can access the cluster by calling kubectl get pods

   nix
   kubectl get pods

   nix
   kubectl get pods

   If your cluster is new it may not have any resources:

   output
   text
   No resources found in default namespace.

   output
   text
   No resources found in default namespace.

Install Jump to heading #

Choose one of the following installation methods:

Install with Helm Jump to heading #

You can install with Helm using either the HTTPS or OCI repository:

This will create an EC2 Classic Load Balancer that routes traffic to the ingress controller service. You can map the load balancer’s IP address to your public DNS domain name.

Install with kubectl Jump to heading #

1. Download the latest source code from the HAProxy Kubernetes Ingress Controller GitHub page.

2. Edit the file deploy/haproxy-ingress.yaml.

3. Change the Service resource in haproxy-ingress.yaml, setting its type field to LoadBalancer, which is the preferred mode when running in the cloud:

   haproxy-ingress.yaml
   yaml
   apiVersion: v1
   kind: Service
   metadata:
     labels:
       run: haproxy-ingress
     name: haproxy-ingress
     namespace: haproxy-controller
   spec:
     selector:
       run: haproxy-ingress
     type: LoadBalancer
     ports:
     - name: http
       port: 80
       protocol: TCP
       targetPort: 80
     - name: https
       port: 443
       protocol: TCP
       targetPort: 443
     - name: stat
       port: 1024
       protocol: TCP
       targetPort: 1024

   haproxy-ingress.yaml
   yaml
   apiVersion: v1
   kind: Service
   metadata:
     labels:
       run: haproxy-ingress
     name: haproxy-ingress
     namespace: haproxy-controller
   spec:
     selector:
       run: haproxy-ingress
     type: LoadBalancer
     ports:
     - name: http
       port: 80
       protocol: TCP
       targetPort: 80
     - name: https
       port: 443
       protocol: TCP
       targetPort: 443
     - name: stat
       port: 1024
       protocol: TCP
       targetPort: 1024

4. Use the kubectl apply command to deploy the controller:

   nix
   kubectl apply -f haproxy-ingress.yaml

   nix
   kubectl apply -f haproxy-ingress.yaml

   This will create an EC2 Classic Load Balancer that routes traffic to the ingress controller service. You can map the load balancer’s IP address to your public DNS domain name, for example by using Route53.

Check your installation Jump to heading #

Verify that the controller is installed into your Kubernetes cluster by using the command kubectl get pods:

nix
kubectl get pods --namespace haproxy-controller

nix
kubectl get pods --namespace haproxy-controller

output
text
NAME                                        READY   STATUS    RESTARTS   AGE
haproxy-kubernetes-ingress-7dd4cc4b-x5fkv   1/1     Running   0          1m

output
text
NAME                                        READY   STATUS    RESTARTS   AGE
haproxy-kubernetes-ingress-7dd4cc4b-x5fkv   1/1     Running   0          1m

Get the External IP, which you can use to access your cluster:

nix
kubectl get services --namespace haproxy-controller

nix
kubectl get services --namespace haproxy-controller

output
text
NAME                         TYPE           CLUSTER-IP       EXTERNAL-IP                                   PORT(S)                                     AGE
haproxy-kubernetes-ingress   LoadBalancer   10.104.173.167   a1234-5678-9012.us-east-2.elb.amazonaws.com   80:30264/TCP,443:31575/TCP,1024:31785/TCP   157m

output
text
NAME                         TYPE           CLUSTER-IP       EXTERNAL-IP                                   PORT(S)                                     AGE
haproxy-kubernetes-ingress   LoadBalancer   10.104.173.167   a1234-5678-9012.us-east-2.elb.amazonaws.com   80:30264/TCP,443:31575/TCP,1024:31785/TCP   157m