Skip to content

Post 4: Installing Longhorn

Published by bobo on May 26, 2025

In the previous post, we set up a basic k3s cluster with MetalLB to handle LoadBalancer IPs. Now it’s time to enable dynamic, persistent storage using Longhorn. Longhorn provides highly available storage across your Kubernetes nodes, making it perfect for configuration files for applications. Local persistent volumes are great for databases as well which struggle in NFS.

Why Longhorn?

Longhorn is a lightweight, distributed block storage system designed specifically for Kubernetes. It works great in smaller setups like ours and supports features like snapshots, backups, and volume replicas across nodes.

Prerequisites

Each k3s node in your cluster should have:

  • A second unformatted disk (we created a 50GB drive for this in Post 2)
  • Ubuntu Server with NFS tools (nfs-common) installed
  • SSH access from the jumpbox

Step-by-Step Installation

Step 1: Label Your Nodes

We’ll label each node to enable scheduling of Longhorn pods:

kubectl label node cube1 longhorn=true
kubectl label node cube2 longhorn=true
kubectl label node cube3 longhorn=true

Step 2: Install Longhorn via Helm

Add the Longhorn Helm repo and install it into its own namespace:

helm repo add longhorn https://charts.longhorn.io
helm repo update

kubectl create namespace longhorn-system

helm install longhorn longhorn/longhorn \
  --namespace longhorn-system \
  --set persistence.defaultClass=true \
  --set defaultSettings.defaultDataPath="/mnt/longhorn"

⚠️ Important: Replace /mnt/longhorn with the mount path of your 50GB disk on each node. You’ll need to manually mount the drive and ensure it persists after reboots.

Step 3: Format and Mount the Second Disk

SSH into each node and run:

sudo mkfs.ext4 /dev/sdb
sudo mkdir -p /mnt/longhorn
sudo mount /dev/sdb /mnt/longhorn

To make it persistent:

echo "/dev/sdb /mnt/longhorn ext4 defaults 0 0" | sudo tee -a /etc/fstab

Step 4: Access the UI

Once deployed, the Longhorn UI should be available. You can expose it using a LoadBalancer service:

apiVersion: v1
kind: Service
metadata:
  name: longhorn-frontend
  namespace: longhorn-system
spec:
  type: LoadBalancer
  loadBalancerIP: 10.0.0.30
  ports:
    - port: 80
      targetPort: 80
  selector:
    app: longhorn-ui

Apply with:

kubectl apply -f longhorn-service.yaml

Then access it at http://10.0.0.30

Step 5: Test It

Create a PersistentVolumeClaim using the longhorn storage class:

apiVersion: v1
kind: PersistentVolumeClaim
metadata:
  name: test-pvc
spec:
  accessModes:
    - ReadWriteOnce
  storageClassName: longhorn
  resources:
    requests:
      storage: 1Gi

Apply with:

kubectl apply -f test-pvc.yaml

You should see the volume show up in the Longhorn dashboard.

With Longhorn up and running, we’re now ready to deploy apps that need persistent storage — like applications, dashboards, and databases. In the next post, we’ll cover setting up NFS access for large media files.

➡️ Stay tuned for Post 5: Configuring NFS Storage for Media Apps.

Published inKubernetes