使用nfs在k8s集群中实现持久化存储

准备NFS服务192.168.1.244
$ yum -y install nfs-utils rpcbind
$ systemctl start nfs-server rpcbind
$ systemctl enable nfs-server rpcbind
$ mkdir -p /data/k8s
$ cd /data/k8s
$ echo 11111111 > index.html
$ vim /etc/exports
/data/k8s *(rw,async,no_root_squash)
$ systemctl restart nfs-server
$ exportfs -arv
客户端测试,所有k8s节点都要安装nfs客户端
$ yum -y install nfs-utils rpcbind
$ systemctl start nfs rpcbind
$ systemctl enable nfs rpcbind
$ showmount -e 192.168.1.244

创建pod直接挂载nfs服务器

apiVersion: v1
kind: Pod
metadata:
  labels:
    run: nginx
  name: podxx
spec:
  volumes:
  - name: nfs
    nfs:
      server: 192.168.1.244
      path: /data/k8s
  containers:
  - image: nginx
    name: nginx
    volumeMounts:
    - mountPath: /usr/share/nginx/html
      name: nfs

$ kubectl exec podxx -it bash

PV 的全称是:PersistentVolume(持久化卷),是对底层的共享存储的一种抽象,PV 由管理员进行创建和配置,它和具体的底层的共享存储技术的实现方式有关,比如 Ceph、GlusterFS、NFS 等,都是通过插件机制完成与共享存储的对接。
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。

Deployment/pod---->pvc---->pv---->共享存储
AccessModes(访问模式)
ReadWriteOnce(RWO):读写权限,但是只能被单个节点挂载
ReadOnlyMany(ROX):只读权限,可以被多个节点挂载
ReadWriteMany(RWX):读写权限,可以被多个节点挂载
persistentVolumeReclaimPolicy(回收策略)
Retain(保留)- 保留数据,需要管理员手工清理数据
Recycle(回收)- 清除 PV 中的数据,效果相当于执行 rm -rf /thevoluem/*
Delete(删除)- 与 PV 相连的后端存储完成 volume 的删除操作,当然这常见于云服务商的存储服务,比如 ASW EBS。
目前只有 NFS 和 HostPath 两种类型支持回收策略。当然一般来说还是设置为 Retain 这种策略保险一点。
一个 PV 的生命周期中,可能会处于4中不同的阶段
Available(可用):表示可用状态,还未被任何 PVC 绑定
Bound(已绑定):表示已经被 PVC 绑定
Released(已释放):PVC 被删除,但是资源还未被集群重新声明
Failed(失败): 表示该 PV 的自动回收失败

手动管理pv和pvc
创建顺序:后端存储---pv---pvc---pod
删除顺序:pod---pvc---pv
1、创建pv

apiVersion: v1
kind: PersistentVolume
metadata:
  name: pv2
  labels:
    app: nfs
spec:
  capacity:
    storage: 1Gi
  accessModes:
  - ReadWriteOnce
  persistentVolumeReclaimPolicy: Recycle
  nfs:
    path: /data/k8s
    server: 192.168.1.244

2、创建pvc

kind: PersistentVolumeClaim
apiVersion: v1
metadata:
  name: pvc2-nfs
spec:
  accessModes:
    - ReadWriteOnce
  resources:
    requests:
      storage: 1Gi
  selector:
    matchLabels:
      app: nfs

上述pvc会自动和具有访问模式是ReadWriteOnce、storage大于等于1Gi、标签是app: nfs的pv进行绑定
3创建pod使用pvc

apiVersion: extensions/v1beta1
kind: Deployment
metadata:
  name: nfs-pvc
spec:
  replicas: 3
  template:
    metadata:
      labels:
        app: nfs-pvc
    spec:
      containers:
      - name: nginx
        image: nginx
        imagePullPolicy: IfNotPresent
        ports:
        - containerPort: 80
          name: web
        volumeMounts:
        - name: www
          subPath: nginx1        #需要手动指定nfs服务器中的子目录,该目录会自动创建
          mountPath: /usr/share/nginx/html
      volumes:
      - name: www
        persistentVolumeClaim:
          claimName: pvc2-nfs        #第2步中创建的pvc

结果是:共享存储中的/data/k8s/nginx1会被挂载到上述pod中的/usr/share/nginx/html目录

使用StorageClass管理pv和pvc
通常情况下,只有pv是动态生成的(使用StatefulSet的pvc模板除外),其他的还是需要手动创建
动态生成pv需要StorageClass和nfs-client-provisioner的共同作用
1、创建nfs-client-provisioner
provisione直接使用nfs服务器
$ docker pull quay.io/external_storage/nfs-client-provisioner:latest

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
            - name: NFS_SERVER
              value: 192.168.1.244
            - name: NFS_PATH
              value: /data/k8s
      volumes:
        - name: nfs-client-root
          nfs:
            server: 192.168.1.244
            path: /data/k8s
---
apiVersion: v1
kind: ServiceAccount
metadata:
  name: nfs-client-provisioner
---
kind: ClusterRole
apiVersion: rbac.authorization.k8s.io/v1
metadata:
  name: nfs-client-provisioner-runner
rules:
  - apiGroups: [""]
    resources: ["persistentvolumes"]
    verbs: ["get", "list", "watch", "create", "delete"]
  - apiGroups: [""]
    resources: ["persistentvolumeclaims"]
    verbs: ["get", "list", "watch", "update"]
  - apiGroups: ["storage.k8s.io"]
    resources: ["storageclasses"]
    verbs: ["get", "list", "watch"]
  - apiGroups: [""]
    resources: ["events"]
    verbs: ["list", "watch", "create", "update", "patch"]
  - apiGroups: [""]
    resources: ["endpoints"]
    verbs: ["create", "delete", "get", "list", "watch", "patch", "update"]
---
kind: ClusterRoleBinding
apiVersion: rbac.authorization.k8s.io/v1
metadata:
  name: run-nfs-client-provisioner
subjects:
  - kind: ServiceAccount
    name: nfs-client-provisioner
    namespace: default
roleRef:
  kind: ClusterRole
  name: nfs-client-provisioner-runner
  apiGroup: rbac.authorization.k8s.io

2、创建StorageClass

apiVersion: storage.k8s.io/v1
kind: StorageClass
metadata:
  name: course-nfs-storage
provisioner: fuseim.pri/ifs

3、创建pvc并动态生成pv

apiVersion: v1
kind: PersistentVolumeClaim
metadata:
  name: test-pvc
  annotations:
    volume.beta.kubernetes.io/storage-class: "course-nfs-storage"
spec:
  accessModes:
  - ReadWriteMany
  resources:
    requests:
      storage: 500Mi

$ kubectl get pv #pv会自动生成,并和pvc绑定

4、使用上述pvc创建pod
和手动管理pv和pvc不同的是,使用StorageClass管理pv和pvc在创建pod时不用手动指定子目录,会在存储服务器根目录/data/k8s内自动生成一个随机子目录

apiVersion: extensions/v1beta1
kind: Deployment
metadata:
  name: nfs-pvc
spec:
  replicas: 3
  template:
    metadata:
      labels:
        app: nfs-pvc
    spec:
      containers:
      - name: nginx
        image: nginx
        imagePullPolicy: IfNotPresent
        ports:
        - containerPort: 80
          name: web
        volumeMounts:
        - name: www
          mountPath: /usr/share/nginx/html
      volumes:
      - name: www
        persistentVolumeClaim:
          claimName: test-pvc

经过测试,不管是手动创建一个pod,还是用deployment创建多个pod,都只能生成一个随机目录,也就是一对pvc和pv对应一个持久存储目录

5、创建StatefulSet时,直接在其pvc模板中使用StorageClas自动生成pv和pvc
$ kubectl explain StatefulSet.spec
volumeClaimTemplates <[]Object>

apiVersion: apps/v1beta1
kind: StatefulSet
metadata:
  name: nfs-web
spec:
  serviceName: "nginx"
  replicas: 2
  template:
    metadata:
      labels:
        app: nfs-web
    spec:
      terminationGracePeriodSeconds: 10
      containers:
      - name: nginx
        image: nginx
        ports:
        - containerPort: 80
          name: web
        volumeMounts:
        - name: www
          mountPath: /usr/share/nginx/html
  volumeClaimTemplates:
  - metadata:
      name: www
      annotations:
        volume.beta.kubernetes.io/storage-class: course-nfs-storage
    spec:
      accessModes: [ "ReadWriteOnce" ]
      resources:
        requests:
          storage: 1Gi

上述脚本会生成两个pvc,两个pv,两个pod,在共享存储根目录/data/k8s内会自动生成两个随机子目录,因为副本数是2
和deployment不同,有状态的应用必须每个pod有一个持久存储文件夹,不能多个pod共享一个

你可能感兴趣的:(使用nfs在k8s集群中实现持久化存储)