kubernetes persistent storage volumes PersistentVolume, persistent storage volumes statement PersistentVolumeClaim use

Prepare a machine to install nfs

#安装nfs-utils
yum -y install nfs-utils
systemctl start nfs
systemctl enable nfs

# Create a directory
mkdir / Data / Volumes -pv
CD / Data / Volumes
mkdir {1,2,3,4,5} V

Configuration File Sharing

vim /etc/exports
/data/volumes/v1 192.168.1.0/24(rw,sync,no_root_squash)
/data/volumes/v2 192.168.1.0/24(rw,sync,no_root_squash)
/data/volumes/v3 192.168.1.0/24(rw,sync,no_root_squash)
/data/volumes/v4 192.168.1.0/24(rw,sync,no_root_squash)
/data/volumes/v5 192.168.1.0/24(rw,sync,no_root_squash)

# Set remount / etc / exports of
exportfs -arv

# Show host's / etc / exports shared directory data
showmount -e

K8s nfs mounted on all nodes node
yum the install nfs -Y-utils
systemctl Start nfs
systemctl enable nfs

Whether nfs functioning test
mkdir / v1
echo the Test v1 >> /v1/test.txt
# 192.168.1.190 is nfs host ip
Mount -t nfs 192.168.1.190:/data/volumes/v1 / v1
echo the Test v1 >> / v1 / test.txt

# Nfs node
cat /data/volumes/v1/test.txt 

test.txt on node node node has been synchronized to nfs, No problem.

# Uninstall / v1 directory on the node test node
umount / v1

 

Creating PersistentVolume, PersistentVolumeClaim, Pod on the master node

vim pv.yaml

apiVersion: v1
kind: PersistentVolume
metadata:
  name: pv1
  labels:
    name: pv1
spec:
  nfs:
# path 挂载路径
    path: /data/volumes/v1
# server nfs机器的ip
    server: 192.168.1.190
# ReadWriteOnce能以读写模式被加载到一个节点上
  accessModes: ["ReadWriteOnce"]
  capacity:
    storage: 2Gi
---
apiVersion: v1
kind: PersistentVolume
metadata:
  name: pv2
  labels:
    name: pv2
spec:
  nfs:
    path: /data/volumes/v2
    server: 192.168.1.190
# ReadOnlyMany以只读模式加载到多个节点上
  accessModes: ["ReadOnlyMany"]
  capacity:
    storage: 5Gi
---
apiVersion: v1
kind: PersistentVolume
metadata:
  name: pv3
  labels:
    name: pv3
spec:
  nfs:
    path: /data/volumes/v3
    server: 192.168.1.190
# ReadWriteMany以读写模式被加载到多个节点上    
  accessModes: ["ReadWriteMany", "ReadWriteOnce"]
  capacity:
    storage: 10Gi
---
apiVersion: v1
kind: PersistentVolume
metadata:
  name: pv4
  labels:
    name: pv4
spec:
  nfs:
    path: /data/volumes/v4
    server: 192.168.1.190
  accessModes: ["ReadWriteMany", "ReadWriteOnce"]
  capacity:
    storage: 10Gi

vim pod-pvc.yaml 

apiVersion: v1
# 存储卷声明
kind: PersistentVolumeClaim
metadata:
  name: pvc-myapp
  namespace: default
spec:
  accessModes: ["ReadWriteMany"]
  resources:
    requests:
      storage: 1Gi
---
apiVersion: v1
kind: Pod
metadata:
  name: pod-myapp
  namespace: default
spec:
  volumes:
  - name: html
#   使用pvc-myapp存储卷声明
    persistentVolumeClaim:
      claimName: pvc-myapp
  containers:
  - name: myapp
    image: ikubernetes/myapp:v1
    volumeMounts:
    - name: html
      mountPath: /usr/share/nginx/html

deploy

kubectl apply -f pv.yaml 
kubectl apply -f pod-pvc.yaml 

View pv

kubectl get pv

View pvc

kubectl get pvc

Use the pv3, nfs node to the new /data/volumes/v3/index.html

echo This is the home page pv3 >> /data/volumes/v3/index.html
 

View pod

kubectl get pods -o wide

curl 10.244.2.39

The nfs / data / volumes / v3 pod container and / usr / share / nginx / html / directory sync

Modify / usr / share / nginx / html / sync to see if the pod container / data / volumes / v3

Exec -it POD-myapp kubectl - / bin / SH
cd / usr / report this content share / nginx / HTML
echo add, modify index.html >>
echo add files >> new.txt

Available to the nfs / data / volumes / v3 to see whether the changes synchronized.

 

Delete pod, pvc

kubectl delete -f pod-pvc.yaml 

kubectl get pv

pv3 state is Released

Released: release state, indicating that PVC unbundling PV, but not yet performed the recovery strategy.
That is, after deleting pvc, pv can not be bound with the use of new pvc.

Pv3 want to use again can be achieved by modifying the definition of pv3

kubectl edit pv pv3

Delete spec.claimRef (the contents of the red box) to

kubectl get pv

Available became a pv.