Setup an NFS client provisioner in Kubernetes

Setup an NFS client provisioner in Kubernetes

One of the most common needs when deploying Kubernetes is the ability to use shared storage. While there are several options available, one of the most commons and easier to setup is to use an NFS server.
This post will explain how to setup a dynamic NFS client provisioner on Kubernetes, relying on an existing NFS server on your systems.

Step 1. Setup an NFS server (sample for CentOS)

First thing you will need, of course, is to have an NFS server. This can be easily achieved with some easy steps:

• Install nfs package: yum install -y nfs-utils
• Enable and start nfs service and rpcbind:
  systemctl enable rpcbind
  systemctl enable nfs-server
  systemctl start rpcbind
  systemctl start nfs-server
  
• Create the directory that will be shared by NFS, and change the permissions:
  mkdir /var/nfsshare
  chmod -R 755 /var/nfsshare
  chown nfsnobody:nfsnobody /var/nfsshare
  
•  Share the NFS directory over the network, creating the /etc/exports file:
  vi /etc/exports
  /var/nfsshare * (rw,sync,no_root_squash,no_all_squash)
• Restart the nfs service to apply the content:
  systemctl restart nfs-server
• Add NFS and rpcbind services to firewall:
  firewall-cmd --permanent --zone=public --add-service=nfs
  firewall-cmd --permanent --zone=public --add-service=rpcbind
  firewall-cmd --reload
  
  The NFS server is now ready to be used

Step 2. Install NFS client provisioner

To achieve that, we will rely on Kubernetes external storage provisioner (https://github.com/kubernetes-incubator/external-storage) . An external provisioner is a dynamic volume provisioner, whose code lives outside kubernetes code.
It relies on an StorageClass object, that defines the external provisioner instance. Then, that instance will wait for PersistentVolumeClaims asking for that specific StorageClass, and will automatically create PersistentVolumes.
In that case we are relying on NFS-client provisioner (https://github.com/kubernetes-incubator/external-storage/tree/master/nfs-client) , that will provide those volumes, relying on an existing NFS server.

In order to use that, several steps are needed:

• Clone the external-storage repository and switch to the nfs-client folder:
  git clone https://github.com/kubernetes-incubator/external-storage
  cd external-storage/nfs-client
• Customize the deploy/class.yaml file, to give a custom provisioner name to your instance:
  apiVersion: storage.k8s.io/v1
  kind: StorageClass
  metadata:
    name: managed-nfs-storage
  provisioner: fuseim.pri/ifs # or choose another name, must match deployment's env PROVISIONER_NAME'
  parameters:
    archiveOnDelete: "false"
• Customize the deploy/deployment.yaml file, to specify the location and folder for your NFS server, and to give the right provisioner name:
  kind: ServiceAccount
  metadata:
    name: nfs-client-provisioner
  ---
  kind: Deployment
  apiVersion: extensions/v1beta1
  metadata:
    name: nfs-client-provisioner
  spec:
    replicas: 1
    strategy:
      type: Recreate
    template:
     metadata:
      labels:
        app: nfs-client-provisioner
     spec:
       serviceAccountName: nfs-client-provisioner
       containers:
         - name: nfs-client-provisioner
           image: quay.io/external_storage/nfs-client-provisioner:latest
           volumeMounts:
             - name: nfs-client-root
               mountPath: /persistentvolumes
           env:
             - name: PROVISIONER_NAME
               value: fuseim.pri/ifs # or choose another name, must match
             - name: NFS_SERVER
               value: <<IP_OF_YOUR_NFS_SERVER>>
             - name: NFS_PATH
               value: <<PATH_TO_NFS_SHARED_FOLDER>>
       volumes:
         - name: nfs-client-root
           nfs:
             server: <<IP_OF_YOUR_NFS_SERVER>>
             path: <<PATH_TO_NFS_SHARED_FOLDER>>
             apiVersion: v1
• Create the objects into your kubernetes cluster:
  kubectl create -f deploy/rbac.yaml
  kubectl create -f deploy/class.yaml
  kubectl create -f deploy/deployment.yaml
• To actually start testing the system, you will need to create a PersistentVolumeClaim using that StorageClass. Then you will need to create a pod that uses this PersistentVolumeClaim:
  claim.yaml
  ----------kind: PersistentVolumeClaim
  apiVersion: v1
  metadata:
    name: test-claim
    annotations:
      volume.beta.kubernetes.io/storage-class: "managed-nfs-storage"
  spec:
    accessModes:
      - ReadWriteMany
    resources:
      requests:
        storage: 1Mi
  
  pod.yaml
  --------
  kind: Pod
  apiVersion: v1
  metadata:
    name: test-pod
  spec:
    containers:
     - name: test-pod
       image: gcr.io/google_containers/busybox:1.24
       command:
         - "/bin/sh"
       args:
         - "-c"
         - "touch /mnt/SUCCESS && exit 0 || exit 1"
       volumeMounts:
         - name: nfs-pvc
           mountPath: "/mnt"
  restartPolicy: "Never"
  volumes:
     - name: nfs-pvc
       persistentVolumeClaim:
         claimName: test-claim

  kubectl create -f deploy/claim.yaml kubectl create -f deploy/pod.yamlYou can see that the pods can be started now with NFS volumes that are dynamically provisioned, relying on your existing NFS server installation.