kubernetes（k8s）PVC

概念

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。

准备工作

在使用 PVC 之前，我们还得把其他节点上的 nfs 客户端给安装上，比如我们这里：

[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

部署PV

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

创建Pod绑定PVC

[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-deploy2.yaml

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 ~]#