[Appendix] Install NFS storage-class

A StorageClass provides a way for administrators to describe the "classes" of storage they offer. Different classes might map to quality-of-service levels, backup policies, or to arbitrary policies determined by the cluster administrators. Kubernetes itself is unopinionated about what classes represent. This concept is sometimes called "profiles" in other storage systems.

Notice: Please configure the following setting in the physical or virtual systems which contain NFS service.

Install NFS related package

For Ubuntu Linux:

sudo apt install -y nfs-common nfs-kernel-server

For RHEL 7/8:

sudo yum install nfs-utils

Configure NFS storage

Add the provisioner and internal IP information to /etc/exports

Notice: The following code needs to be replaced with real-case information.
```
sudo vim /etc/exports
```
```
/provisioner <your_class_c_domain>/16(rw,no_root_squash,no_subtree_check)
```
- Example for /etc/exports
Create the folder to mount and store all files.
```
sudo mkdir -p /provisioner
```

Restart NFS service

sudo exportfs -a
# For Ubuntu
sudo systemctl restart nfs-kernel-server
# For RHEL
sudo systemctl status rpcbind
sudo systemctl restart nfs
sudo systemctl enable nfs

Notice: Please configure the following setting in other physical or virtual systems. You can use the ansible tool to set in the environment.

[Verify] Check the testing file correctly mount into the NFS system.
- Create a file in any node which can mount via os level.
```
mkdir /tmp/testmount
sudo mount -t nfs 10.40.0.15:/provisioner /tmp/testmount/
echo -e "test123\\nline2" | sudo tee /tmp/testmount/test.txt
```
- Check the file is in the NFS mounted folder.
```
cat /provisioner/test.txt
test123
line2
```
- If you finish testing the file, you must need to umount the test mount point
```
sudo umount /tmp/testmount
```

Use helm tool to install nfs storage plugin into Kubernetes system.

Add the helm repo of nfs-subdir-external-provisioner

helm repo add nfs-subdir-external-provisioner <https://kubernetes-sigs.github.io/nfs-subdir-external-provisioner>

helm repo update

Install provisioner by helm

helm install \\
nfs-subdir-external-provisioner nfs-subdir-external-provisioner/nfs-subdir-external-provisioner \\
--create-namespace \\
--namespace nfs \\
--set nfs.server=<nfs-service-node-internal-ip> \\
--set nfs.path=<nfs-mount-path> \\
--set nfs.mountOptions='{nfsvers=4.1}' \\
--set storageClass.reclaimPolicy=Retain \\
--set storageClass.volumeBindingMode=Immediate

Example:

[Verify] the NFS Provisioner status in Kubernetes

Check nfs namespace has been created

kubectl get ns | grep nfs
nfs               Active   54s

Check storage class nfs-client has been created.

kubectl get sc

NAME         PROVISIONER                                     RECLAIMPOLICY   VOLUMEBINDINGMODE   ALLOWVOLUMEEXPANSION   AGE
nfs-client   cluster.local/nfs-subdir-external-provisioner   Retain          Immediate           true                   3m52s

Make the NFS StorageClass as the default storage class

kubectl annotate sc nfs-client storageclass.kubernetes.io/is-default-class=true
kubectl get sc | grep default

NAME                   PROVISIONER                                     RECLAIMPOLICY   VOLUMEBINDINGMODE   ALLOWVOLUMEEXPANSION   AGE
nfs-client (default)   cluster.local/nfs-subdir-external-provisioner   Retain          Immediate           true                   5m48s

[Verify] Test client storage can take mount point /provisioner.

Create nfs-client-pvc pods

kubectl -n nfs apply -f - <<EOF
apiVersion: v1
kind: Pod
metadata:
  name: nfs-client-pvc
spec:
  containers:
  - command: ['/bin/sh', '-c']
    args: ["sleep 999d"]
    image: busybox
    name: main
    resources:
      limits:
        cpu: "50m"
        memory: 128Mi
    volumeMounts:
    - name: data
      mountPath: /**provisioner**
  volumes:
  - name: data
    persistentVolumeClaim:
      claimName: nfs-client-pvc
---
apiVersion: v1
kind: PersistentVolumeClaim
metadata:
  name: nfs-client-pvc
spec:
  accessModes:
  - ReadWriteOnce
  resources:
    requests:
      storage: 2Gi
  storageClassName: nfs-client
  volumeMode: Filesystem
EOF

Check the nfs-client-pvc pod is running.

kubectl -n nfs get pod | grep nfs-client-pvc
nfs-client-pvc                                    1/1     Running   0          48s

Check the PVC which STATUS column of PVC should Bound.

kubectl -n nfs get pvc | grep nfs-client-pvc
NAME                STATUS   VOLUME                                     CAPACITY   ACCESS MODES   STORAGECLASS   AGE
nfs-client-pvc      **Bound**    pvc-8b64019d-6aed-4349-b8ba-5aaaede6a4ee   2Gi        RWO            nfs-client     6m27s

Execute onto pod shell in the master node which contains kubectl tool and create files in nfs-client-pvc pods
```
kubectl -n nfs exec -it nfs-client-pvc -- sh
```

In nfs-client-pvc pods, we will create the testing file which can be seen in nfs storage.

<pod> $ stat /provisioner/
  File: /provisioner/
  Size: 4096      	Blocks: 8          IO Block: 32768  directory
<pod> $ echo "test123" | tee /provisioner/test.txt
test123
<pod> $ exit

Check the testing file in the /provisioner folder in the NFS node.
```
cat /provisioner/test.txt
test123
line2
```
If you can get the file in the NFS node, you can delete the nfs-client-pvc testing pod.
```
kubectl -n nfs delete pod nfs-client-pvc
pod "nfs-client-pvc" deleted
```