PVC 的全称是:PersistentVolumeClaim(持久化卷声明),PVC 是用户存储的一种声明,PVC 和 Pod 比较类似,Pod 消耗的是节点,PVC 消耗的是 PV 资源,Pod 可以请求 CPU 和内存,而 PVC 可以请求特定的存储空间和访问模式。对于真正使用存储的用户不需要关心底层的存储实现细节,只需要直接使用 PVC
即可。
本文中有一个master,3个node节点
192.168.8.81 k8master1.meng.com k8master1 k8kubapi.meng.com k8kubapi
192.168.8.84 k8node1.meng.com k8node1
192.168.8.85 k8node2.meng.com k8node2
192.168.8.86 k8node3.meng.com k8node3
查看master和node节点的状态
[root@k8master1 ~]# kubectl get nodes
NAME STATUS ROLES AGE VERSION
k8master1.meng.com Ready control-plane 22d v1.28.1
k8node1.meng.com Ready <none> 22d v1.28.1
k8node2.meng.com Ready <none> 22d v1.28.1
k8node3.meng.com Ready <none> 22d v1.28.1
需要在所有节点安装 nfs 客户端程序,必须在所有节点都安装 nfs 客户端,否则可能会导致 PV 挂载不上的问题。
安装命令如下:
yum -y install nfs-utils rpcbind
systemctl start rpcbind.service
systemctl start nfs.service
在master节点创建一些文件夹,更改属主
mkdir -p /nfs1/data/01 /nfs1/data/02 /nfs1/data/03
chown -R nfsnobody.nfsnobody /nfs1/
将新建的文件夹共享
[root@k8master1 ~]# vim /etc/exports
添加如下内容
/nfs1/data/01 192.168.8.0/24(rw,sync,root_squash,all_squash)
/nfs1/data/02 192.168.8.0/24(rw,sync,root_squash,all_squash)
/nfs1/data/03 192.168.8.0/24(rw,sync,root_squash,all_squash)
从node节点尝试访问master节点共享的NFS文件夹,所有的节点都需要测试一下
[root@k8node1 containerd]# showmount -e 192.168.8.81
Export list for 192.168.8.81:
/nfs1/data/03 192.168.8.0/24
/nfs1/data/02 192.168.8.0/24
/nfs1/data/01 192.168.8.0/24
[root@k8node1 containerd]# mount -t nfs 192.168.8.81:/nfs1/data/01 /mnt
[root@k8node1 /]# umount /mnt
创建了3个PV,大小为10M,1G,3G
[root@k8master1 ~]# cat pv.yaml
apiVersion: v1
kind: PersistentVolume
metadata:
name: pv01-10m
spec:
capacity:
storage: 10M
accessModes:
- ReadWriteMany
storageClassName: nfs
nfs:
path: /nfs1/data/01
server: 192.168.8.81
---
apiVersion: v1
kind: PersistentVolume
metadata:
name: pv02-1gi
spec:
capacity:
storage: 1Gi
accessModes:
- ReadWriteMany
storageClassName: nfs
nfs:
path: /nfs1/data/02
server: 192.168.8.81
---
apiVersion: v1
kind: PersistentVolume
metadata:
name: pv03-3gi
spec:
capacity:
storage: 3Gi
accessModes:
- ReadWriteMany
storageClassName: nfs
nfs:
path: /nfs1/data/03
server: 192.168.8.81
创建PV
[root@k8master1 ~]# kubectl apply -f pv.yaml
查看PV状态
[root@k8master1 ~]# kubectl get pv
NAME CAPACITY ACCESS MODES RECLAIM POLICY STATUS CLAIM STORAGECLASS REASON AGE
pv01-10m 10M RWX Retain Available nfs 4s
pv02-1gi 1Gi RWX Retain Available nfs 4s
pv03-3gi 3Gi RWX Retain Available nfs 4s
创建PVC
[root@k8master1 ~]# cat pvc.yaml
kind: PersistentVolumeClaim
apiVersion: v1
metadata:
name: nginx-pvc
spec:
accessModes:
- ReadWriteMany
resources:
requests:
storage: 200Mi
storageClassName: nfs
运维YAML文件创建PVC
[root@k8master1 ~]# kubectl apply -f pvc.yaml
查看PV,已有一个PV处于绑定状态
[root@k8master1 ~]# kubectl get pv
NAME CAPACITY ACCESS MODES RECLAIM POLICY STATUS CLAIM STORAGECLASS REASON AGE
pv01-10m 10M RWX Retain Available nfs 42h
pv02-1gi 1Gi RWX Retain Bound default/nginx-pvc nfs 42h
pv03-3gi 3Gi RWX Retain Available nfs 42h
发现这个pv02-1gi绑定了我们刚才的pvc,为什么是这个,因为我们申请的是200M,01太小,03太大,k8s帮我们自动做了选择
[root@k8master1 ~]# cat pvc-deploy2.yaml
apiVersion: apps/v1
kind: Deployment
metadata:
labels:
app: nginx-deploy-pvc
name: nginx-deploy-pvc
spec:
replicas: 2
selector:
matchLabels:
app: nginx-deploy-pvc
template:
metadata:
labels:
app: nginx-deploy-pvc
spec:
containers:
- image: nginx
name: nginx
volumeMounts:
- name: html
mountPath: /usr/share/nginx/html
volumes:
- name: html
persistentVolumeClaim:
claimName: nginx-pvc
---
apiVersion: v1
kind: Service
metadata:
name: nginx-pvc
labels:
app: nginx-pvc
spec:
type: NodePort
ports:
- port: 80
targetPort: web #容器端口或名字
selector:
app: nginx-pvc
[root@k8master1 ~]#
执行文件创建pod,创建deployment和service,并暴露给node
[root@k8master1 ~]# kubectl apply -f pvc-deploy.yaml
apiVersion: apps/v1
kind: Deployment
metadata:
labels:
app: nginx-deploy-pvc
name: nginx-deploy-pvc
spec:
replicas: 2
selector:
matchLabels:
app: nginx-deploy-pvc
template:
metadata:
labels:
app: nginx-deploy-pvc
spec:
containers:
- image: nginx
name: nginx
ports:
- containerPort: 80
name: web
volumeMounts:
- name: www
mountPath: /usr/share/nginx/html
volumes:
- name: www
persistentVolumeClaim:
claimName: nginx-pvc
---
apiVersion: v1
kind: Service
metadata:
name: nginx-pvc
labels:
app: nginx-pvc
spec:
type: NodePort
ports:
- port: 80
targetPort: 80
nodePort: 30004
selector:
app: nginx-deploy-pvc
我们进入到02里面新建个文件
[root@k8master1 ~]# cd /nfs1/data/02
[root@k8master1 ~]# echo 111 > index.html
交互访问docker,可以看到新创建的index文件
[root@k8master1 02]# kubectl exec -it nginx-deploy-pvc-7c466b8668-cfpsg -- bash
root@nginx-deploy-pvc-7c466b8668-cfpsg:/# cd /usr/share/nginx/html/
root@nginx-deploy-pvc-7c466b8668-cfpsg:/usr/share/nginx/html# ls
index.html
root@nginx-deploy-pvc-7c466b8668-cfpsg:/usr/share/nginx/html#
创建的pod可以curl一下看看是否可以访问,如果不能访问可以参考K8S遇到的问题的文章。
[root@k8master1 ~]# kubectl get pods -o wide
NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES
demoapp-7c58cd6bb-7qlrd 1/1 Running 1 (16m ago) 22d 10.244.1.15 k8node1.meng.com <none> <none>
demoapp-7c58cd6bb-f84kp 1/1 Running 1 (4m50s ago) 22d 10.244.3.12 k8node3.meng.com <none> <none>
demoapp-7c58cd6bb-ldrzf 1/1 Running 1 (8m50s ago) 22d 10.244.2.13 k8node2.meng.com <none> <none>
mypod 1/1 Running 1 (8m50s ago) 21d 10.244.2.12 k8node2.meng.com <none> <none>
nginx-deploy-pvc-64b6b6bb47-csrnr 1/1 Running 1 (4m50s ago) 20m 10.244.3.11 k8node3.meng.com <none> <none>
nginx-deploy-pvc-64b6b6bb47-wcsbc 1/1 Running 1 (16m ago) 20m 10.244.1.16 k8node1.meng.com <none> <none>
[root@k8master1 ~]# curl 10.244.1.16
111
[root@k8master1 ~]#
通过浏览器可以访问暴露的端口来访问Nginx了,master节点和node节点都可以
kubectl create deploy nginx-deploy-pvc --image=nginx:latest --port=80 --replicas=2
kubectl expose deploy nginx-deploy-pvc --name=nginx-pvc --type=NodePort --port=80 --target-port=80