# Self-hosting A Kubernetes Cluster using K3S

> This guide covers setup on a 5-node cluster, from installation to managing with Helm and Portainer, including an Nginx deployment example.

By Zsolt Bizderi · Published 2024-07-29
Canonical: https://ambientnode.uk/self-hosting-a-kubernetes-cluster-using-k3s

[K3S](https://k3s.io/) is a lightweight Kubernetes distribution developed by Rancher Labs, designed to run on resource-constrained environments such as edge devices, IoT appliances, and even low-powered servers. This guide walks you through the basic steps to deploy K3S on a cluster of 5 machines.

## Prerequisites

1. **Operating System**: Ensure all nodes are running a supported OS (e.g., Ubuntu, Debian, CentOS).
2. **User Permissions**: You need root or sudo privileges on all nodes.
3. **Network Configuration**: Ensure that all nodes can communicate with each other over the network.
4. **5 Virtual Machines:** This example uses 5 nodes in a cluster (3 server, 2 worker nodes). You can [use a template](/proxmox-vm-deployment-automation/) to spin up the machines in a few minutes.

### Step 1: Install K3s on the First Master Node

* **Install K3s on Master Node 1:**
  + `curl -sfL` [`https://get.k3s.io`](https://get.k3s.io) `| sh -`
* **Retrieve the Node Token:**
  + `cat /var/lib/rancher/k3s/server/node-token`
* **Get the K3s Master Node 1 IP Address.**
  + `hostname -I`

The token and the IP address are used to join other nodes to the cluster.

### Step 2: Install K3s on the Other Master Nodes

* **Install K3s on Master Nodes 2 and 3:**
  + SSH into Master Node 2 and run the following command to join the node to the cluster (replace `<MASTER_NODE_1_IP>` and `<NODE_TOKEN>` with the actual IP address and token):

```
curl -sfL https://get.k3s.io | K3S_URL=https://<MASTER_NODE_1_IP>:6443 K3S_TOKEN=<NODE_TOKEN> sh -
```

+ Repeat the above step for Master Node 3.

* **Verify Connection on the Master Node 1**:
  + `kubectl get nodes`
* **Set the Roles on Master Node 1**:
  + `kubectl label node MASTER_NODE_2_NAME node-role.kubernetes.io/master=control-plane`
  + `kubectl label node MASTER_NODE_3_NAME node-role.kubernetes.io/master=control-plane`

### Step 3: Install K3S on the Worker Nodes

* **Install K3s on Worker Nodes 1 and 2:**
  + SSH into Worker Node 1 and run the following command to join the node to the cluster (replace `<MASTER_NODE_1_IP>` and `<NODE_TOKEN>` with the actual IP address and token):

```
curl -sfL https://get.k3s.io | K3S_URL=https://<MASTER_NODE_1_IP>:6443 K3S_TOKEN=<NODE_TOKEN> sh -
```

+ Repeat the above step for Worker Node 2.

* **Verify Connection on the Master Node 1**:
  + `kubectl get nodes`
* **Set the Roles on Master Node 1**:
  + `kubectl label node WORKER_NODE_1_NAME node-role.kubernetes.io/worker=worker`
  + `kubectl label node WORKER_NODE_2_NAME node-role.kubernetes.io/worker=worker`

### Step 4: Verify the Cluster

* **Verify the Cluster Status**:
  + `kubectl get nodes`

### Step 5: Install Helm

* Download the Most Recent Helm Binary Executable:
  + `curl -fsSL -o get_helm.sh https://raw.githubusercontent.com/helm/helm/master/scripts/get-helm-3`
* Add Execute Permissions to the Downloaded Script and Run It:
  + `chmod 700 get_helm.sh`
  + `./get_helm.sh`

### Step 6: Deploy Portainer

* **Deploy Portainer**:
  + `sudo kubectl apply -n portainer -f https://raw.githubusercontent.com/portainer/k8s/master/deploy/manifests/portainer/portainer.yaml`
* **Access the Portainer Web Interface**:
  + Open a web browser and go to `http://<NODE_IP>:30777`, where `<NODE_IP>` is the IP address of any of your K3S nodes.
  + Follow the on-screen instructions to set up the admin user and password.
  + Add your K3S cluster to Portainer, it will be automatically recognised.

### Example Deployment: Nginx

* **Create a Deployment File on Master Node 1:**

```
apiVersion: apps/v1
kind: Deployment
metadata:
  name: nginx-deployment
spec:
  replicas: 3
  selector:
    matchLabels:
      app: nginx
  template:
    metadata:
      labels:
        app: nginx
    spec:
      containers:
      - name: nginx
        image: nginx:latest
        ports:
        - containerPort: 80
---
apiVersion: v1
kind: Service
metadata:
  name: nginx-service
spec:
  type: LoadBalancer
  selector:
    app: nginx
  ports:
    - protocol: TCP
      port: 80
      targetPort: 80
```

nginx-deployment.yaml

* **Apply the Deployment**:
  + `kubectl apply -f nginx-deployment.yaml`
* **Check the Status of Your Pods**:
  + `kubectl get pods`
* **Check the Status of the Service to Get the External IP or NodePort**:
  + `kubectl get services`
* **Describe the Service to Get the NodePort**:
  + `kubectl describe service nginx-service`
* **Access the Service**:
  + Open your browser and navigate to the IP address of any of your K3S nodes using the NodePort obtained.
* **Clean Up the Resources after testing**:
  + `kubectl delete -f nginx-deployment.yaml`

This guide provides a structured approach to deploying a K3S cluster and managing it with Helm and Portainer, complete with an example Nginx deployment.