K3S: Storage Management

Understanding Persistent Volumes (PV) and Persistent Volume Claims (PVC)#

Storage in Kubernetes is like keeping track of your socks in the laundry—things disappear unless you manage them properly. Persistent Volumes (PV) ensure your data survives beyond the lifespan of a Pod.

What are Persistent Volumes (PV) and why are they needed?#

A PV is a piece of storage allocated for a cluster, while a Persistent Volume Claim (PVC) is how applications request storage. Think of it as Kubernetes’ way of making sure storage isn’t just a free-for-all.

Difference between PV and PVC#

• PV: The actual storage resource, defined by the administrator.
• PVC: A request for storage by applications, which Kubernetes tries to match with an available PV.

Example YAML for creating a Persistent Volume#

apiVersion: v1
kind: PersistentVolume
metadata:
  name: my-pv
spec:
  capacity:
    storage: 5Gi
  accessModes:
    - ReadWriteOnce
  persistentVolumeReclaimPolicy: Retain
  hostPath:
    path: "/mnt/data"

Creating a Persistent Volume Claim (PVC)#

apiVersion: v1
kind: PersistentVolumeClaim
metadata:
  name: my-pvc
spec:
  accessModes:
    - ReadWriteOnce
  resources:
    requests:
      storage: 2Gi

Applying PV and PVC#

kubectl apply -f my-pv.yaml
kubectl apply -f my-pvc.yaml
kubectl get pv,pvc

Using Different Storage Classes in K3s#

Understanding Storage Classes in Kubernetes#

StorageClasses allow dynamic provisioning of storage. Instead of pre-allocating a PV, Kubernetes can create one on-demand.

Listing available storage classes#

kubectl get storageclass

Defining a custom StorageClass#

apiVersion: storage.k8s.io/v1
kind: StorageClass
metadata:
  name: fast-storage
provisioner: rancher.io/local-path
volumeBindingMode: WaitForFirstConsumer

Assigning a StorageClass to a PVC#

apiVersion: v1
kind: PersistentVolumeClaim
metadata:
  name: my-fast-pvc
spec:
  storageClassName: fast-storage
  accessModes:
    - ReadWriteOnce
  resources:
    requests:
      storage: 5Gi

Applying the StorageClass and PVC#

kubectl apply -f fast-storage.yaml
kubectl apply -f my-fast-pvc.yaml

Local Storage vs. Cloud Storage Options#

Local Storage#

• Uses the host machine for storage (hostPath, localPath)
• Best for on-prem clusters, but doesn’t scale well

Cloud Storage Providers#

• AWS EBS, Google Persistent Disk, Azure Disk provide managed storage
• Example AWS EBS StorageClass:

apiVersion: storage.k8s.io/v1
kind: StorageClass
metadata:
  name: aws-ebs
provisioner: kubernetes.io/aws-ebs
parameters:
  type: gp2
  fsType: ext4

Implementing NFS and Longhorn Storage in K3s#

Using NFS as a Storage Provider#

NFS (Network File System) allows multiple nodes to share the same storage, making it perfect for multi-node K3s clusters.

Setting up an NFS-based Persistent Volume#

apiVersion: v1
kind: PersistentVolume
metadata:
  name: nfs-pv
spec:
  capacity:
    storage: 10Gi
  accessModes:
    - ReadWriteMany
  nfs:
    server: 192.168.1.100
    path: "/data"

Using Longhorn for Dynamic Storage in K3s#

Longhorn is a lightweight, distributed block storage system for Kubernetes, designed for simplicity and high availability.

Installing Longhorn in K3s#

kubectl apply -f https://raw.githubusercontent.com/longhorn/longhorn/master/deploy/longhorn.yaml

Verifying Longhorn installation#

kubectl get pods -n longhorn-system

Creating a Longhorn-based StorageClass#

apiVersion: storage.k8s.io/v1
kind: StorageClass
metadata:
  name: longhorn
provisioner: driver.longhorn.io

Hands-On Exercise#

Now it’s time to put all this storage knowledge into action:

• Set up a Persistent Volume (PV) and Persistent Volume Claim (PVC) in K3s
• Configure and deploy an application with persistent storage
• Install and test Longhorn storage in K3s
• Use NFS for multi-node storage across a K3s cluster

With these skills, you’ll be managing Kubernetes storage like a pro—no more lost data, no more missing socks. 🚀