Kubernetes使用cephfs作为后端存储

这里使用了k8s自身的持久化卷存储机制：PV和PVC。各组件之间的关系参考下图：

 

PV的Access Mode（访问模式）

The access modes are:

    ReadWriteOnce – the volume can be mounted as read-write by a single node

    ReadOnlyMany – the volume can be mounted read-only by many nodes

    ReadWriteMany – the volume can be mounted as read-write by many nodes

In the CLI, the access modes are abbreviated to:

    RWO - ReadWriteOnce

    ROX - ReadOnlyMany

    RWX - ReadWriteMany

Reclaim Policy（回收策略）

Current reclaim policies are:

    Retain – manual reclamation

    Recycle – basic scrub (“rm -rf /thevolume/*”)

    Delete – associated storage asset such as AWS EBS, GCE PD, Azure Disk, or OpenStack Cinder volume is deleted

Currently, only NFS and HostPath support recycling. AWS EBS, GCE PD, Azure Disk, and Cinder volumes support deletion.

PV是有状态的资源对象，有以下几种状态：

1、Available：空闲状态

2、Bound：已经绑定到某个pvc上

3、Released：对应的pvc已经删除，但资源还没有被回收

4、Failed：pv自动回收失败

1、创建secret。在secret中，data域的各子域的值必须为base64编码值。

#echo "AQDchXhYTtjwHBAAk2/H1Ypa23WxKv4jA1NFWw==" | base64
QVFEY2hYaFlUdGp3SEJBQWsyL0gxWXBhMjNXeEt2NGpBMU5GV3c9PQo=

#vim ceph-secret.yaml
apiVersion: v1
kind: Secret
metadata:
  name: ceph-secret
data:
  key: QVFEY2hYaFlUdGp3SEJBQWsyL0gxWXBhMjNXeEt2NGpBMU5GV3c9PQo=

2、创建pv。pv只能是网络存储，不属于任何node，但可以在每个node上访问。pv并不是定义在pod上的，而是独立定义于pod之外。目前pv仅支持定义存储容量。

#vim ceph-pv.yml
apiVersion: v1
kind: PersistentVolume
metadata:
  name: cephfs
spec:
  capacity:
    storage: 10Gi
  accessModes:
    - ReadWriteMany
  fsType: xfs
  cephfs:
    monitors:
    - 172.16.100.5 :6789
    - 172.16.100.6 :6789
    - 172.16.100.7 :6789
    path: /opt/eshop_dir/eshop
    user: admin
    secretRef:
      name: ceph-secret

3、创建pvc

#vim ceph-pvc.yml
kind: PersistentVolumeClaim
apiVersion: v1
metadata:
  name: cephfs
spec:
  accessModes:
      - ReadWriteMany
  resources:
      requests:
        storage: 8Gi

4、查看pv和pvc

#kubectl get pv
cephfs      10Gi       RWX           Retain          Bound     default/cephfs           2d
#kubectl get pvc
cephfs      Bound     cephfs         10Gi       RWX           2d

5、创建rc，这个只是测试样例

#vim ceph-rc.yml
kind: ReplicationController
metadata: 
  name: cephfstest
  labels: 
    name: cephfstest
spec: 
  replicas: 4
  selector:
    name: cephfstest
  template:
    metadata:
      labels:
        name: cephfstest
    spec: 
      containers: 
      - name: cephfstest
        image: 172.60.0.107/pingpw/nginx-php:v4
        env:
        - name: GET_HOSTS_FROM
          value: env
        ports: 
        - containerPort: 81
        volumeMounts:
        - name: cephfs
          mountPath: "/opt/cephfstest"
      volumes:
      - name: cephfs
        persistentVolumeClaim:
           claimName: cephfs

5、查看pod

#kubectl get pod -o wide
cephfstest-09j37                            1/1       Running   0          2d        10.244.5.16   kuber-node03
cephfstest-216r6                            1/1       Running   0          2d        10.244.3.25   kuber-node01
cephfstest-4sjgr                            1/1       Running   0          2d        10.244.4.26   kuber-node02
cephfstest-p2x7c                            1/1       Running   0          2d        10.244.6.22   kuber-node04