NFS Storage Share in Kubernetes

In this blog, we will explore in detail how to present the NFS file mount to a Kubernetes cluster running on an Ubuntu operating system. NFS (Network File System) is a popular file-sharing protocol that enables a user to access files located on remote servers as if they were stored locally.

Mounting an NFS file system to a Kubernetes cluster provides several benefits, including increased flexibility and scalability. In this post, we will go through the steps required to mount an NFS file system to a Kubernetes cluster running on Ubuntu. We will cover the prerequisites, the configuration steps required to install and set up the NFS server and client, and finally, the steps to mount the NFS share to the Kubernetes cluster.

By the end of this post, you should have a clear understanding of how to set up an NFS file system and present it to a Kubernetes cluster running on Ubuntu. So, let’s dive in and get started!

There are two ways to access data via NFS within Kubernetes:

Persistent Volume with NFS: this lets you set up a managed resource within the cluster that is accessed via NFS.

Directly map the NFS volume to the pod: Kubernetes allows you to mount a Volume as a local drive on a container.

Step1: Persistent volume with NFS

Requirement: NFS Server

NFS Server configuraion: https://medium.com/cloudnloud/nfs-server-and-client-configuration-62a7e9c8bad8

Diagram

PV Creation

root@Kubernet-Master:~# cat pv.yaml
apiVersion: v1
kind: PersistentVolume
metadata:
  name: nfs-pv
  labels:
    name: mynfs # name can be anything
spec:
  storageClassName: manual # same storage class as pvc
  capacity:
    storage: 200Mi
  accessModes:
    - ReadWriteMany
  nfs:
    server: 192.168.100.102 # NFS server IP address 
    path: "/mnt/nfs_share"   # NFS share location 
root@Kubernet-Master:~#

PV Status

root@Kubernet-Master:~# kubectl get pv
NAME     CAPACITY   ACCESS MODES   RECLAIM POLICY   STATUS   CLAIM             STORAGECLASS   REASON   AGE
nfs-pv   200Mi      RWX            Retain           Bound    default/nfs-pvc   manual                  41h

PVC creation

root@Kubernet-Master:~# cat pvc.yaml
apiVersion: v1
kind: PersistentVolumeClaim
metadata:
  name: nfs-pvc
spec:
  storageClassName: manual
  accessModes:
    - ReadWriteMany #  must be the same as PersistentVolume
  resources:
    requests:
      storage: 50Mi

PVC status

root@Kubernet-Master:~# kubectl get pvc
NAME      STATUS   VOLUME   CAPACITY   ACCESS MODES   STORAGECLASS   AGE
nfs-pvc   Bound    nfs-pv   200Mi      RWX            manual         41h
root@Kubernet-Master:~#

Crating the sample nginx deployment with PVC

apiVersion: apps/v1
kind: Deployment
metadata:
  labels:
    app: nginx
  name: nfs-nginx
spec:
  replicas: 1
  selector:
    matchLabels:
      app: nginx
  template:
    metadata:
      labels:
        app: nginx
    spec:
      volumes:
      - name: nfs-test
        persistentVolumeClaim:
          claimName: nfs-pvc # same name of pvc that was created
      containers:
      - image: nginx
        name: nginx
        volumeMounts:
        - name: nfs-test # name of volume should match claimName volume
          mountPath: /usr/share/nginx/html # mount inside of contianer

Creating the deployment

root@Kubernet-Master:~# kubectl create -f deployment.yaml
deployment.apps/nfs-nginx created

Testing the pod status

root@Kubernet-Master:~# kubectl get pod
NAME                         READY   STATUS    RESTARTS        AGE
nfs-nginx-5c89779fb6-v8lnv   1/1     Running   0               59m

Testing the volume status

root@Kubernet-Master:~# kubectl exec -it nfs-nginx-5c89779fb6-v8lnv bash
kubectl exec [POD] [COMMAND] is DEPRECATED and will be removed in a future version. Use kubectl exec [POD] -- [COMMAND] instead.
root@nfs-nginx-5c89779fb6-v8lnv:/# df -h
Filesystem                      Size  Used Avail Use% Mounted on
overlay                          98G   13G   80G  14% /
tmpfs                            64M     0   64M   0% /dev
tmpfs                           957M     0  957M   0% /sys/fs/cgroup
shm                              64M     0   64M   0% /dev/shm
/dev/sda5                        98G   13G   80G  14% /etc/hosts
192.168.100.102:/mnt/nfs_share   98G   13G   80G  14% /usr/share/nginx/html
tmpfs                           1.8G   12K  1.8G   1% /run/secrets/kubernetes.io/serviceaccount
tmpfs                           957M     0  957M   0% /proc/acpi
tmpfs                           957M     0  957M   0% /proc/scsi
tmpfs                           957M     0  957M   0% /sys/firmware

Step2 Directly map the NFS volume to the pod

Diagram

Pod creation manifest file

In the same file, mention the NFS server details and mount location

kind: Pod
apiVersion: v1
metadata:
  name: nfs
spec:
  containers:
    - name: app
      image: httpd
      volumeMounts:
        - name: nfs-volume
          mountPath: /var/nfs # Please change the destination you like the share to be mounted too
  volumes:
    - name: nfs-volume
      nfs:
        server: 192.168.100.102  # NFS Server Ip address 
        path: /mnt/nfs_share  # NFS Server share location

Creating the pod

root@Kubernet-Master:~# kubectl create -f NFSPOD.yaml
pod/nfs created

POD Status

root@Kubernet-Master:~# kubectl get pod
NAME   READY   STATUS    RESTARTS   AGE
nfs    1/1     Running   0          82s
root@Kubernet-Master:~#

Verifying the mount point

root@Kubernet-Master:~# kubectl exec -it nfs bash
kubectl exec [POD] [COMMAND] is DEPRECATED and will be removed in a future version. Use kubectl exec [POD] -- [COMMAND] instead.
root@nfs:/usr/local/apache2# df -h
Filesystem                      Size  Used Avail Use% Mounted on
overlay                          98G   13G   80G  14% /
tmpfs                            64M     0   64M   0% /dev
tmpfs                           957M     0  957M   0% /sys/fs/cgroup
192.168.100.102:/mnt/nfs_share   98G   13G   80G  14% /var/nfs
dev/sda5                        98G   13G   80G  14% /etc/hosts
shm                              64M     0   64M   0% /dev/shm
tmpfs                           1.8G   12K  1.8G   1% /run/secrets/kubernetes.io/serviceaccount
tmpfs                           957M     0  957M   0% /proc/acpi
tmpfs                           957M     0  957M   0% /proc/scsi
tmpfs                           957M     0  957M   0% /sys/firmware
root@nfs:/usr/local/apache2#

By following the steps, you should now have a good understanding of how to configure and use NFS as a persistent storage option in your Kubernetes applications. We hope this post has been informative and helpful.

Happy Learning 📚

Thank you!