本文主要是介绍K8s 使用 Ceph RBD 作为后端存储（静态供给、动态供给），希望对大家解决编程问题提供一定的参考价值，需要的开发者们随着小编来一起学习吧！
一、K8s 使用 Ceph RBD
Ceph RBD（Rados Block Device）是 Ceph 存储集群中的一个重要组件，它提供了块级别的存储访问。RBD 允许用户创建虚拟块设备，并将其映射到客户端系统中，就像本地磁盘一样使用。
首先所有 k8s 节点都需要安装 ceph-common 工具：
yum -y install epel-release ceph-common
二、静态供给方式
静态供给方式需要提前创建好 RBD Image 给到 K8s 使用。
2.1 在 Ceph 中创建 RBD Image 和 授权用户
RBD 需要绑定一个存储池，创建一个名为 rbd-static-pool 的存储池：
ceph osd pool create rbd-static-pool 8
指定存储池的特性为 RBD，并初始化该存储池给 RBD 使用：
# 指定类型
ceph osd pool application enable rbd-static-pool rbd
# 初始化
rbd pool init -p rbd-static-pool
创建 RBD Image：
# 创建
rbd create rbd-static-volume -s 5G -p rbd-static-pool
# 关闭特征 ，格式 {pool name}/{image name}
rbd feature disable rbd-static-pool/rbd-static-volume object-map fast-diff deep-flatten
创建用户 rbd-static-user 并授权存储池 rbd-static-pool
ceph auth get-or-create client.rbd-static-user mon 'allow r' osd 'allow class-read object_prefix rbd_children, allow rwx pool=rbd-static-pool' -o ceph.client.rbd-static-user.keyring
查看 rbd-static-user 用户的秘钥：
ceph auth get-key client.rbd-static-user
2.2 k8s 创建 secret 配置上面的 key
export RBD_STATIC_USER_SECRET='AQA1xyBmIxD/LxAAByBhiw68kiiWKclYyjfU8A=='kubectl create secret generic rbd-static-user-secret --type="kubernetes.io/rbd" \
--from-literal=key=$RBD_STATIC_USER_SECRET \
--namespace=default
2.3 pod 直接使用 RBD 存储
vi rbd-test-pod.yml
apiVersion: v1
kind: Pod
metadata:name: rbd-test-pod
spec:containers:- name: nginximage: nginximagePullPolicy: IfNotPresentvolumeMounts:- name: data-volumemountPath: /usr/share/nginx/html/volumes:- name: data-volumerbd:monitors: ["11.0.1.140:6789"]pool: rbd-static-poolimage: rbd-static-volumeuser: rbd-static-usersecretRef:name: rbd-static-user-secretfsType: ext4
kubectl apply -f rbd-test-pod.yml
查看 pod ：
kubectl get pods
可以进到 pod 中查看磁盘情况：
kubectl exec -it rbd-test-pod -- /bin/bashlsblk
2.4 创建 PV 使用 RBD 存储
vi rbd-test-pv.yml
apiVersion: v1
kind: PersistentVolume
metadata:name: rbd-test-pv
spec:accessModes: ["ReadWriteOnce"]capacity:storage: 2GipersistentVolumeReclaimPolicy: Retainrbd:monitors: ["11.0.1.140:6789"]pool: rbd-static-poolimage: rbd-static-volumeuser: rbd-static-usersecretRef:name: rbd-static-user-secretfsType: ext4
kubectl apply -f rbd-test-pv.yml
创建 PVC 绑定 PV ：
vi rbd-test-pvc.yml
apiVersion: v1
kind: PersistentVolumeClaim
metadata:name: rbd-test-pvc
spec:accessModes: ["ReadWriteOnce"]resources:requests:storage: 2Gi
kubectl apply -f rbd-test-pvc.yml
查看 pvc 和 pv ：
kubectl get pvckubectl get pv
测试 pod 挂载 pvc：
vi rbd-test-pod1.yml
apiVersion: v1
kind: Pod
metadata:name: rbd-test-pod1
spec:containers:- name: nginximage: nginximagePullPolicy: IfNotPresentvolumeMounts:- name: data-volumemountPath: /usr/share/nginx/html/volumes:- name: data-volumepersistentVolumeClaim:claimName: rbd-test-pvcreadOnly: false
kubectl apply -f rbd-test-pod1.yml
查看 pod ：
kubectl get pods
三、动态供给方式
使用动态存储时 controller-manager 需要使用 rbd 命令创建 image，但是官方controller-manager镜像里没有rbd命令，所以在开始前需要安装 rbd-provisioner 插件：
vi external-storage-rbd-provisioner.yml
apiVersion: v1
kind: ServiceAccount
metadata:name: rbd-provisionernamespace: 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: ServiceAccountname: rbd-provisionernamespace: kube-system
roleRef:kind: ClusterRolename: rbd-provisionerapiGroup: rbac.authorization.k8s.io---
apiVersion: rbac.authorization.k8s.io/v1
kind: Role
metadata:name: rbd-provisionernamespace: kube-system
rules:
- apiGroups: [""]resources: ["secrets"]verbs: ["get"]
---
apiVersion: rbac.authorization.k8s.io/v1
kind: RoleBinding
metadata:name: rbd-provisionernamespace: kube-system
roleRef:apiGroup: rbac.authorization.k8s.iokind: Rolename: rbd-provisioner
subjects:
- kind: ServiceAccountname: rbd-provisionernamespace: kube-system
---
apiVersion: apps/v1
kind: Deployment
metadata:name: rbd-provisionernamespace: kube-system
spec:selector:matchLabels:app: rbd-provisionerreplicas: 1strategy:type: Recreatetemplate:metadata:labels:app: rbd-provisionerspec:containers:- name: rbd-provisionerimage: "registry.cn-chengdu.aliyuncs.com/ives/rbd-provisioner:v2.0.0-k8s1.11"env:- name: PROVISIONER_NAMEvalue: ceph.com/rbdserviceAccount: rbd-provisioner
kubectl apply -f external-storage-rbd-provisioner.yml
查看 pod：
kubectl get pods -n kube-system
3.1 在 ceph 创建存储池 和 授权用户
由于 k8s 会自动创建 rbd 这就仅仅创建存储池即可：
ceph osd pool create dynamics-pool 8
创建 k8s 访问用户：
ceph auth get-or-create client.dynamics-user mon 'allow r' osd 'allow class-read object_prefix rbd_children, allow rwx pool=dynamics-pool' -o ceph.client.dynamics-user.keyring
查看 dynamics-user 的秘钥：
ceph auth get-key client.dynamics-user
k8s 动态供给还需要 admin 用户的秘钥：
ceph auth get-key client.admin
3.2 在 k8s 中创建 secret
export ADMIN_USER_SECRET='AQDdUB9mcTI3LRAAOBpt3e7AH5v9fiMtHKQpqA=='kubectl create secret generic ceph-admin-secret --type="kubernetes.io/rbd" \
--from-literal=key=$ADMIN_USER_SECRET \
--namespace=kube-systemexport DYNAMICS_USER_SECRET='AQAtuyBmi3w6AhAADEU3mKkFRSqDyRLs46aSXA=='kubectl create secret generic ceph-dynamics-user-secret --type="kubernetes.io/rbd" \
--from-literal=key=$DYNAMICS_USER_SECRET \
--namespace=default
3.3 创建 StorageClass
vi ceph-sc.yml
kind: StorageClass
apiVersion: storage.k8s.io/v1
metadata:name: dynamic-ceph-rdb
provisioner: ceph.com/rbd
parameters:monitors: 11.0.1.140:6789adminId: adminadminSecretName: ceph-admin-secretadminSecretNamespace: kube-systempool: dynamics-pooluserId: dynamics-useruserSecretName: ceph-dynamics-user-secretfsType: ext4imageFormat: "2"imageFeatures: "layering"
kubectl apply -f ceph-sc.yml
查看 SC：
kubectl get sc
3.4 测试创建 PVC
vi test-pvc.yml
kind: PersistentVolumeClaim
apiVersion: v1
metadata:name: rbd-pvc1
spec:accessModes: - ReadWriteOncestorageClassName: dynamic-ceph-rdbresources:requests:storage: 2Gi
kubectl apply -f test-pvc.yml
查看 pvc：
kubectl get pvc
创建 pod 使用上面 pvc ：
vi rbd-test-pod2.yml
apiVersion: v1
kind: Pod
metadata:name: rbd-test-pod2
spec:containers:- name: nginximage: nginximagePullPolicy: IfNotPresentvolumeMounts:- name: data-volumemountPath: /usr/share/nginx/html/volumes:- name: data-volumepersistentVolumeClaim:claimName: rbd-pvc1readOnly: false
kubectl apply -f rbd-test-pod2.yml
查看 pod ：
kubectl get pods
3.5 测试使用 volumeClaimTemplates 动态创建 pv 和 pvc
vi mysql.yml
# headless service
apiVersion: v1
kind: Service
metadata:name: mysql-hlnamespace: mysqllabels:app: mysql-hl
spec:clusterIP: Noneports:- name: mysql-portport: 3306selector:app: mysql---
# NodePort service
apiVersion: v1
kind: Service
metadata:name: mysql-npnamespace: mysqllabels:app: mysql-np
spec:clusterIP: ports:- name: master-portport: 3306nodePort: 31306targetPort: 3306selector:app: mysqltype: NodePorttarget-port:externalTrafficPolicy: Cluster ---
apiVersion: apps/v1
kind: StatefulSet
metadata:name: mysql
spec:serviceName: "mysql-hl"replicas: 1selector: matchLabels: app: mysqltemplate:metadata:labels:app: mysqlspec:containers:- name: mysqlimage: mysql:8.0.20ports:- containerPort: 3306name: master-portenv:- name: MYSQL_ROOT_PASSWORDvalue: "root"- name: TZvalue: "Asia/Shanghai"volumeMounts: - name: mysql-datamountPath: /var/lib/mysql volumeClaimTemplates:- metadata:name: mysql-dataspec:accessModes: ["ReadWriteOnce"]storageClassName: dynamic-ceph-rdbresources:requests:storage: 2Gi
kubectl apply -f mysql.yml
查看 pod ：
查看 pvc ：
查看 pv：
这篇关于K8s 使用 Ceph RBD 作为后端存储（静态供给、动态供给）的文章就介绍到这儿，希望我们推荐的文章对编程师们有所帮助！
