The goal of this project is to create a HomeLab Server to deploy our own applications using a cluster of Raspberry Pi with Kubernetes.

Materials

• Two Rasbperry Pi (ideally Rpi4 with 4Gb RAM at least for master node)
• Wifi or Ethernet connection
• Two SD Cards

Master node installation

1. Install the SO (see Appendix A)
2. Put the SD Card in the Rasbperry Pi and turn on it
3. Access to the node with ssh

   1
   ssh master01@master01.local
   

4. Modify the file /boot/cmdline.txt with sudo adding next value. Edit the ip address value for your case. In my case I decided to put the static ip address 192.168.1.200

   1
   cgroup_memory=1 cgroup_enable=memory ip=192.168.1.200::192.168.1.1:255.255.255.0:master01.local:eth0:off
   

5. Restart the server and access again with ssh

   1
   sudo reboot
   

   1
   ssh master01@master01.local
   

6. Install k3s server. Replace the static address with your value.

   1
   curl -sfL https://get.k3s.io | sh -s - --bind-address 192.168.1.200 --write-kubeconfig-mode 644
   

7. Check that the cluster is running correctly

   1
   2
   3
   4
   > kubectl cluster-info
   Kubernetes control plane is running at https://192.168.1.200:6443
   CoreDNS is running at https://192.168.1.200:6443/api/v1/namespaces/kube-system/services/kube-dns:dns/proxy
   Metrics-server is running at https://192.168.1.200:6443/api/v1/namespaces/kube-system/services/https:metrics-server:https/proxy
   

8. Copy the value of node-token necessary to install the Agent

   1
   sudo cat /var/lib/rancher/k3s/server/node-token
   

Agent node installation

1. Install the SO (see Appendix A)
2. Put the SD Card in the Rasbperry Pi and turn on it
3. Access to the node with ssh

   1
   ssh agent01@agent01.local
   

4. Modify the file /boot/cmdline.txt with sudo adding next value. Edit the ip address value for your case. In my case I decided to put the static ip address 192.168.1.200

   1
   cgroup_memory=1 cgroup_enable=memory
   

5. Restart the server and access again with ssh

   1
   sudo reboot
   

   1
   ssh agent01@agent01.local
   

6. Go to master node with ssh and copy the noden-token value required for the connection between the Agent and the Master

   1
   ssh master01@master01.local
   

   1
   sudo cat /var/lib/rancher/k3s/server/node-token
   

7. Install k3s agent and replace the ip and the value of the token

   1
   curl -sfL https://get.k3s.io | K3S_URL=https://192.168.1.200:6443 K3S_TOKEN=XXXXXX sh -
   

8. Test the agent in the master node

   1
   2
   3
   4
   > master01@master01:~ $ k3s kubectl get nodes
   NAME       STATUS   ROLES                  AGE   VERSION
   master01   Ready    control-plane,master   57m   v1.24.3+k3s1
   agent01    Ready    <none>                 20s   v1.24.3+k3s1
   

Access to the cluster

Configure kubectl from your client to access to the cluster

1. Install kubectl
2. Copy the configuration from master node

   1
   sudo cat /etc/rancher/k3s/k3s.yaml
   

3. Create a file ~/.kube/config and paste the content replacing the values from master node

   1
   2
   3
   4
   5
   6
   7
   8
   9
   10
   11
   12
   13
   14
   15
   16
   17
   18
   apiVersion: v1
   kind: Config
   clusters:
     - cluster:
      server: "https://192.168.1.200:6443"
      certificate-authority-data: LS0....
    name: k3s-cluster
   contexts:
     - context:
      cluster: "k3s-cluster"
      user: "k3s-admin"
    name: k3s
    current-context: k3s
   users:
     - name: k3s-admin
    user:
      client-certificate-data: LS0t...
      client-key-data: LS0t...
   

4. Invoke use context and check the connection

   1
   kubectl config use-context k3s
   

   1
   2
   3
   4
   > kubectl get nodes
   NAME       STATUS   ROLES                  AGE   VERSION
   agent01    Ready    <none>                 23m   v1.24.3+k3s1
   master01   Ready    control-plane,master   81m   v1.24.3+k3s1
   

Example: Install Home Assistant

Installation:

1
2
3
helm repo add k8s-at-home https://k8s-at-home.com/charts/
helm repo update
helm install home-assistant k8s-at-home/home-assistant

Now with a port-forward:

1
kubectl port-forward deployment/home-assistant 8123:8123

you can check your installation in your browser localhost:8123

Uninstallation:

1
helm uninstall home-assistant

More info in Helm home-assistant

Appendix A: How to install Raspberry Pi OS

• Install Raspberry Pi OS using Raspberry Pi Imager in the SD Card http://raspberrypi.com/
• Use Raspberry Pi OS Lite (64 bits)
• Advanced options:
  • Set hostname. Example: “master01.local” or “agent01.local” depending on the type of node
  • Enable SSH using password authentication
  • Define user/password. Example: master01 or agent01
  • Set locale settings
  • Wifi is your connection is not using ethernet

Appendix B: Uninstall K3S

Master

1
/usr/local/bin/k3s-uninstall.sh

Agent

1
/usr/local/bin/k3s-agent-uninstall.sh