k8s使用ceph实现动态持久化存储

简介

本文章介绍如何使用ceph为k8s提供动态申请pv的功能。ceph提供底层存储功能，cephfs方式支持k8s的pv的3种访问模式ReadWriteOnce，ReadOnlyMany ，ReadWriteMany ，RBD支持ReadWriteOnce，ReadOnlyMany两种模式

访问模式只是能力描述，并不是强制执行的，对于没有按pvc声明的方式使用pv，存储提供者应该负责访问时的运行错误。例如如果设置pvc的访问模式为ReadOnlyMany ，pod挂载后依然可写，如果需要真正的不可写，申请pvc是需要指定 readOnly: true 参数

部署

部署k8s

centos7使用kubeadm安装k8s-1.11版本

部署ceph

centos7安装ceph分布式存储集群

在k8s集群中配置使用ceph

使用Ceph RBD

使用kubeadm安装集群的额外配置

# 如果使用kubeadm部署的集群需要这些额外的步骤
# 由于使用动态存储时 controller-manager 需要使用 rbd 命令创建 image
# 所以 controller-manager 需要使用 rbd 命令
# 由于官方controller-manager镜像里没有rbd命令
# 如果没使用如下方式会报错无法成功创建pvc
# 相关 issue https://github.com/kubernetes/kubernetes/issues/38923
cat >external-storage-rbd-provisioner.yaml<<EOF
apiVersion: v1
kind: ServiceAccount
metadata:
  name: rbd-provisioner
  namespace: kube-system
---
kind: ClusterRole
apiVersion: rbac.authorization.k8s.io/v1
metadata:
  name: rbd-provisioner
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: ["create", "update", "patch"]
  - apiGroups: [""]
    resources: ["endpoints"]
    verbs: ["get", "list", "watch", "create", "update", "patch"]
  - apiGroups: [""]
    resources: ["services"]
    resourceNames: ["kube-dns"]
    verbs: ["list", "get"]
---
kind: ClusterRoleBinding
apiVersion: rbac.authorization.k8s.io/v1
metadata:
  name: rbd-provisioner
subjects:
  - kind: ServiceAccount
    name: rbd-provisioner
    namespace: kube-system
roleRef:
  kind: ClusterRole
  name: rbd-provisioner
  apiGroup: rbac.authorization.k8s.io

---
apiVersion: rbac.authorization.k8s.io/v1
kind: Role
metadata:
  name: rbd-provisioner
  namespace: kube-system
rules:
- apiGroups: [""]
  resources: ["secrets"]
  verbs: ["get"]
---
apiVersion: rbac.authorization.k8s.io/v1
kind: RoleBinding
metadata:
  name: rbd-provisioner
  namespace: kube-system
roleRef:
  apiGroup: rbac.authorization.k8s.io
  kind: Role
  name: rbd-provisioner
subjects:
- kind: ServiceAccount
  name: rbd-provisioner
  namespace: kube-system

---
apiVersion: extensions/v1beta1
kind: Deployment
metadata:
  name: rbd-provisioner
  namespace: kube-system
spec:
  replicas: 1
  strategy:
    type: Recreate
  template:
    metadata:
      labels:
        app: rbd-provisioner
    spec:
      containers:
      - name: rbd-provisioner
        image: "quay.io/external_storage/rbd-provisioner:v2.0.0-k8s1.11"
        env:
        - name: PROVISIONER_NAME
          value: ceph.com/rbd
      serviceAccount: rbd-provisioner
EOF
kubectl apply -f external-storage-rbd-provisioner.yaml

# 查看状态 等待running之后 再进行后续的操作
kubectl get pod -n kube-system

配置 storageclass

# 在k8s集群中所有节点安装 ceph-common
# 需要使用kubelet使用rdb命令map附加rbd创建的image
yum install -y ceph-common

# 创建 osd pool 在ceph的mon或者admin节点
ceph osd pool create kube 4096
ceph osd pool ls

# 创建k8s访问ceph的用户 在ceph的mon或者admin节点
ceph auth get-or-create client.kube mon 'allow r' osd 'allow class-read object_prefix rbd_children, allow rwx pool=kube' -o ceph.client.kube.keyring

# 查看key 在ceph的mon或者admin节点
ceph auth get-key client.admin
ceph auth get-key client.kube

# 创建 admin secret
# CEPH_ADMIN_SECRET 替换为 client.admin 获取到的key
export CEPH_ADMIN_SECRET='AQBBAnRbSiSOFxAAEZXNMzYV6hsceccYLhzdWw=='
kubectl create secret generic ceph-secret --type="kubernetes.io/rbd" \
--from-literal=key=$CEPH_ADMIN_SECRET \
--namespace=kube-system

# 在 default 命名空间创建pvc用于访问ceph的 secret
# CEPH_KUBE_SECRET 替换为 client.kube 获取到的key
export CEPH_KUBE_SECRET='AQBZK3VbTN/QOBAAIYi6CRLQcVevW5HM8lunOg=='
kubectl create secret generic ceph-user-secret --type="kubernetes.io/rbd" \
--from-literal=key=$CEPH_KUBE_SECRET \
--namespace=default

# 查看 secret
kubectl get secret ceph-user-secret -o yaml
kubectl get secret ceph-secret -n kube-system -o yaml

# 配置 StorageClass
# 如果使用kubeadm创建的集群 provisioner 使用如下方式
# provisioner: ceph.com/rbd
cat >storageclass-ceph-rdb.yaml<<EOF
kind: StorageClass
apiVersion: storage.k8s.io/v1
metadata:
  name: dynamic-ceph-rdb
provisioner: ceph.com/rbd
# provisioner: kubernetes.io/rbd
parameters:
  monitors: 11.11.11.111:6789,11.11.11.112:6789,11.11.11.113:6789
  adminId: admin
  adminSecretName: ceph-secret
  adminSecretNamespace: kube-system
  pool: kube
  userId: kube
  userSecretName: ceph-user-secret
  fsType: ext4
  imageFormat: "2"
  imageFeatures: "layering"
EOF

# 创建
kubectl apply -f storageclass-ceph-rdb.yaml

# 查看
kubectl get sc

测试使用

# 创建pvc测试
cat >ceph-rdb-pvc-test.yaml<<EOF
kind: PersistentVolumeClaim
apiVersion: v1
metadata:
  name: ceph-rdb-claim
spec:
  accessModes:     
    - ReadWriteOnce
  storageClassName: dynamic-ceph-rdb
  resources:
    requests:
      storage: 2Gi
EOF
kubectl apply -f ceph-rdb-pvc-test.yaml
 
# 查看
kubectl get pvc
kubectl get pv
 
# 创建 nginx pod 挂载测试
cat >nginx-pod.yaml<<EOF
apiVersion: v1
kind: Pod
metadata:
  name: nginx-pod1
  labels:
    name: nginx-pod1
spec:
  containers:
  - name: nginx-pod1
    image: nginx:alpine
    ports:
    - name: web
      containerPort: 80
    volumeMounts:
    - name: ceph-rdb
      mountPath: /usr/share/nginx/html
  volumes:
  - name: ceph-rdb
    persistentVolumeClaim:
      claimName: ceph-rdb-claim
EOF
kubectl apply -f nginx-pod.yaml
 
# 查看
kubectl get pods -o wide
 
# 修改文件内容
kubectl exec -ti nginx-pod1 -- /bin/sh -c 'echo Hello World from Ceph RBD!!! > /usr/share/nginx/html/index.html'
 
# 访问测试
POD_ID=$(kubectl get pods -o wide | grep nginx-pod1 | awk '{print $(NF-1)}')
curl http://$POD_ID

# 清理
kubectl delete -f nginx-pod.yaml
kubectl delete -f ceph-rdb-pvc-test.yaml

使用 CephFS

linux内核需要4.10+，否则会出现无法正常使用的问题，详细issue信息 https://github.com/kubernetes-incubator/external-storage/issues/345
centos7升级内核

在ceph集群创建CephFS

# 如下操作在ceph的mon或者admin节点

# CephFS需要使用两个Pool来分别存储数据和元数据
ceph osd pool create fs_data 128
ceph osd pool create fs_metadata 128
ceph osd lspools

# 创建一个CephFS
ceph fs new cephfs fs_metadata fs_data

# 查看
ceph fs ls

部署cephfs-provisioner

# 官方没有cephfs动态卷支持
# 使用社区提供的cephfs-provisioner
cat >external-storage-cephfs-provisioner.yaml<<EOF
apiVersion: v1
kind: ServiceAccount
metadata:
  name: cephfs-provisioner
  namespace: kube-system
---
kind: ClusterRole
apiVersion: rbac.authorization.k8s.io/v1
metadata:
  name: cephfs-provisioner
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: ["create", "update", "patch"]
  - apiGroups: [""]
    resources: ["endpoints"]
    verbs: ["get", "list", "watch", "create", "update", "patch"]
  - apiGroups: [""]
    resources: ["secrets"]
    verbs: ["create", "get", "delete"]
---
kind: ClusterRoleBinding
apiVersion: rbac.authorization.k8s.io/v1
metadata:
  name: cephfs-provisioner
subjects:
  - kind: ServiceAccount
    name: cephfs-provisioner
    namespace: kube-system
roleRef:
  kind: ClusterRole
  name: cephfs-provisioner
  apiGroup: rbac.authorization.k8s.io

---
apiVersion: rbac.authorization.k8s.io/v1
kind: Role
metadata:
  name: cephfs-provisioner
  namespace: kube-system
rules:
  - apiGroups: [""]
    resources: ["secrets"]
    verbs: ["create", "get", "delete"]
---
apiVersion: rbac.authorization.k8s.io/v1
kind: RoleBinding
metadata:
  name: cephfs-provisioner
  namespace: kube-system
roleRef:
  apiGroup: rbac.authorization.k8s.io
  kind: Role
  name: cephfs-provisioner
subjects:
- kind: ServiceAccount
  name: cephfs-provisioner
  namespace: kube-system

---
apiVersion: extensions/v1beta1
kind: Deployment
metadata:
  name: cephfs-provisioner
  namespace: kube-system
spec:
  replicas: 1
  strategy:
    type: Recreate
  template:
    metadata:
      labels:
        app: cephfs-provisioner
    spec:
      containers:
      - name: cephfs-provisioner
        image: "quay.io/external_storage/cephfs-provisioner:v2.0.0-k8s1.11"
        env:
        - name: PROVISIONER_NAME
          value: ceph.com/cephfs
        command:
        - "/usr/local/bin/cephfs-provisioner"
        args:
        - "-id=cephfs-provisioner-1"
      serviceAccount: cephfs-provisioner
EOF
kubectl apply -f external-storage-cephfs-provisioner.yaml

# 查看状态 等待running之后 再进行后续的操作
kubectl get pod -n kube-system

配置 storageclass

# 查看key 在ceph的mon或者admin节点
ceph auth get-key client.admin

# 创建 admin secret
# CEPH_ADMIN_SECRET 替换为 client.admin 获取到的key
# 如果在测试 ceph rbd 方式已经添加 可以略过此步骤
export CEPH_ADMIN_SECRET='AQBBAnRbSiSOFxAAEZXNMzYV6hsceccYLhzdWw=='
kubectl create secret generic ceph-secret --type="kubernetes.io/rbd" \
--from-literal=key=$CEPH_ADMIN_SECRET \
--namespace=kube-system

# 查看 secret
kubectl get secret ceph-secret -n kube-system -o yaml

# 配置 StorageClass
cat >storageclass-cephfs.yaml<<EOF
kind: StorageClass
apiVersion: storage.k8s.io/v1
metadata:
  name: dynamic-cephfs
provisioner: ceph.com/cephfs
parameters:
    monitors: 11.11.11.111:6789,11.11.11.112:6789,11.11.11.113:6789
    adminId: admin
    adminSecretName: ceph-secret
    adminSecretNamespace: "kube-system"
    claimRoot: /volumes/kubernetes
EOF

# 创建
kubectl apply -f storageclass-cephfs.yaml

# 查看
kubectl get sc

测试使用

# 创建pvc测试
cat >cephfs-pvc-test.yaml<<EOF
kind: PersistentVolumeClaim
apiVersion: v1
metadata:
  name: cephfs-claim
spec:
  accessModes:     
    - ReadWriteOnce
  storageClassName: dynamic-cephfs
  resources:
    requests:
      storage: 2Gi
EOF
kubectl apply -f cephfs-pvc-test.yaml
 
# 查看
kubectl get pvc
kubectl get pv
 
# 创建 nginx pod 挂载测试
cat >nginx-pod.yaml<<EOF
apiVersion: v1
kind: Pod
metadata:
  name: nginx-pod1
  labels:
    name: nginx-pod1
spec:
  containers:
  - name: nginx-pod1
    image: nginx:alpine
    ports:
    - name: web
      containerPort: 80
    volumeMounts:
    - name: cephfs
      mountPath: /usr/share/nginx/html
  volumes:
  - name: cephfs
    persistentVolumeClaim:
      claimName: cephfs-claim
EOF
kubectl apply -f nginx-pod.yaml
 
# 查看
kubectl get pods -o wide
 
# 修改文件内容
kubectl exec -ti nginx-pod1 -- /bin/sh -c 'echo Hello World from CephFS!!! > /usr/share/nginx/html/index.html'
 
# 访问测试
POD_ID=$(kubectl get pods -o wide | grep nginx-pod1 | awk '{print $(NF-1)}')
curl http://$POD_ID

# 清理
kubectl delete -f nginx-pod.yaml
kubectl delete -f cephfs-pvc-test.yaml

参考文档

• https://kubernetes.io/docs/concepts/storage/storage-classes/
• https://docs.openshift.com/container-platform/3.5/install_config/storage_examples/ceph_rbd_dynamic_example.html
• https://ieevee.com/tech/2018/05/17/k8s-cephfs.html
• https://github.com/kubernetes-incubator/external-storage/tree/master/ceph/rbd
• https://github.com/kubernetes-incubator/external-storage/blob/master/ceph/rbd/deploy/README.md
• https://github.com/heketi/heketi/blob/master/docs/admin/install-kubernetes.md
• https://github.com/gluster/gluster-kubernetes/blob/master/docs/setup-guide.md
• https://github.com/gluster/gluster-kubernetes/blob/master/docs/examples/hello_world/README.md
• https://jimmysong.io/kubernetes-handbook/practice/using-heketi-gluster-for-persistent-storage.html
• https://kubernetes.io/docs/concepts/storage/persistent-volumes/
• https://docs.openshift.com/enterprise/3.1/architecture/additional_concepts/storage.html#pv-access-modes