Duck GPT-4o

Setting up a home lab with k3s using Alpine Linux and the specified configuration is a great way to learn about Kubernetes and container orchestration. Below is a recommended setup for your Kubernetes cluster, including steps for installation, configuration, and deployment of your applications.

Prerequisites

Physical Machines: Ensure you have two physical machines with virtualization support enabled in the BIOS.
Virtual Machines: Create three VMs:
- VM1: Master Node (k3s server)
- VM2: Worker Node (k3s agent)
- VM3: Worker Node (k3s agent)
Alpine Linux: Install Alpine Linux on all three VMs.
Networking: Ensure that all VMs can communicate with each other over the network.

Step 1: Disable IPv6

To disable IPv6 on Alpine Linux, you can modify the /etc/sysctl.conf file:

echo "net.ipv6.conf.all.disable_ipv6 = 1" >> /etc/sysctl.conf
echo "net.ipv6.conf.default.disable_ipv6 = 1" >> /etc/sysctl.conf
sysctl -p

Step 2: Install k3s

On VM1 (Master Node):
- Install k3s using the following command:
```
curl -sfL https://get.k3s.io | sh -
```
- After installation, retrieve the k3s token for worker nodes:
```
cat /var/lib/rancher/k3s/server/node-token
```
On VM2 and VM3 (Worker Nodes):
- Install k3s as a worker node using the token from VM1:
```
curl -sfL https://get.k3s.io | K3S_URL=https://<VM1_IP>:6443 K3S_TOKEN=<TOKEN> sh -
```
Replace <VM1_IP> with the IP address of VM1 and <TOKEN> with the token you retrieved earlier.

Step 3: Configure NFS for Persistent Storage

On Physical Machine 1:
- Ensure that NFS is installed and configured. Export the desired directory in /etc/exports:
```
/path/to/nfs/share *(rw,sync,no_subtree_check)
```
- Restart the NFS service:
```
exportfs -a
systemctl restart nfs-server
```

On VMs:

Install NFS client:

apk add nfs-utils

Create a mount point and mount the NFS share:

mkdir -p /mnt/nfs
mount -o nolock <Physical_Machine_1_IP>:/path/to/nfs/share /mnt/nfs

To make the mount persistent, add the following line to /etc/fstab:

<Physical_Machine_1_IP>:/path/to/nfs/share /mnt/nfs nfs defaults 0 0

Step 4: Deploy Applications

You can deploy your applications using Helm or Kubernetes manifests. Below are example manifests for each application:

Nextcloud:

Create a nextcloud-deployment.yaml file:

apiVersion: apps/v1
kind: Deployment
metadata:
  name: nextcloud
spec:
  replicas: 1
  selector:
    matchLabels:
      app: nextcloud
  template:
    metadata:
      labels:
        app: nextcloud
    spec:
      containers:
      - name: nextcloud
        image: nextcloud
        ports:
        - containerPort: 80
        volumeMounts:
        - name: nextcloud-storage
          mountPath: /var/www/html
      volumes:
      - name: nextcloud-storage
        nfs:
          server: <Physical_Machine_1_IP>
          path: /path/to/nfs/share/nextcloud

Vaultwarden:

Create a vaultwarden-deployment.yaml file:


apiVersion: apps/v1
kind: Deployment
metadata:
  name: vaultwarden
spec:
  replicas: 1
  selector:
    matchLabels:
      app: vaultwarden
  template:
    metadata:
      labels:
        app: vaultwarden
    spec:
      containers:
      - name: vaultwarden
        image: vaultwarden/server
        ports:
        - containerPort: 80
        volumeMounts:
        - name: vaultwarden-storage
          mountPath: /data
      volumes:
      - name: vaultwarden-storage