k8s使用rbd作为存储

如果需要使用rbd作为后端存储的话，需要先安装ceph-common

1. ceph集群创建rbd

需要提前在ceph集群上创建pool，然后创建image

[root@ceph01 ~]# ceph osd pool create pool01
[root@ceph01 ~]# ceph osd pool application enable pool01 rbd
[root@ceph01 ~]# rbd pool init pool01
[root@ceph01 ~]# rbd create pool01/test --size 10G --image-format 2 --image-feature layerin
[root@ceph01 ~]# rbd info pool01/test

2. k8s编写yaml文件

apiVersion: apps/v1
kind: Deployment
metadata:
  creationTimestamp: null
  labels:
    app: rbd
  name: rbd
spec:
  replicas: 1
  selector:
    matchLabels:
      app: rbd
  strategy: {}
  template:
    metadata:
      creationTimestamp: null
      labels:
        app: rbd
    spec:
      volumes:
      - name: test
        rbd:
          fsType: xfs
          keyring: /root/admin.keyring
          monitors:
          - 192.168.200.230:6789
          pool: pool01
          image: test 
          user: admin
          readOnly: false
      containers:
      - image: nginx
        imagePullPolicy: IfNotPresent
        volumeMounts:
        - mountPath: /usr/share/nginx/html
          name: test
        name: nginx
        resources: {}
status: {}

[root@master ~]# kubectl get pods
NAME                         READY   STATUS              RESTARTS        AGE
rbd-888b8b747-n56wr          1/1     Running   0               26m

这个时候k8s就使用了rbd作为存储

如果这个地方一直显示ContainerCreating的话，可能是没有安装ceph-common,也可能是你的keyring或者ceph.conf没有发放到node节点，具体可以使用describe来看

2.1 进入容器查看挂载

[root@master euler]# kubectl exec -it rbd-5db4759c-nj2b4 -- bash
root@rbd-5db4759c-nj2b4:/# df -hT |grep /dev/rbd0
/dev/rbd0      xfs       10G  105M  9.9G   2% /usr/share/nginx/html

可以看到，/dev/rbd0已经被格式化成xfs并且挂载到了/usr/share/nginx/html

2.2 进入容器修改内容

root@rbd-5db4759c-nj2b4:/usr/share/nginx# cd html/
root@rbd-5db4759c-nj2b4:/usr/share/nginx/html# ls
root@rbd-5db4759c-nj2b4:/usr/share/nginx/html# echo 123 > index.html
root@rbd-5db4759c-nj2b4:/usr/share/nginx/html# chmod 644 index.html
root@rbd-5db4759c-nj2b4:/usr/share/nginx/html# exit
[root@master euler]# kubectl get pods -o wide
NAME                 READY   STATUS    RESTARTS   AGE    IP                NODE    NOMINATED NODE   READINESS GATES
rbd-5db4759c-nj2b4   1/1     Running   0          8m5s   192.168.166.131   node1   <none>           <none>

访问容器查看内容

[root@master euler]# curl 192.168.166.131
123

内容可以正常被访问到，我们将容器删除，然后让他自己重新启动一个来看看文件是否还存在

[root@master euler]# kubectl delete pods rbd-5db4759c-nj2b4 
pod "rbd-5db4759c-nj2b4" deleted
[root@master euler]# kubectl get pods
NAME                 READY   STATUS              RESTARTS   AGE
rbd-5db4759c-v9cgm   0/1     ContainerCreating   0          2s
[root@master euler]# kubectl get pods -owide
NAME                 READY   STATUS    RESTARTS   AGE   IP                NODE    NOMINATED NODE   READINESS GATES
rbd-5db4759c-v9cgm   1/1     Running   0          40s   192.168.166.132   node1   <none>           <none>
[root@master euler]# curl 192.168.166.132
123

可以看到，也是没有问题的，这样k8s就正常的使用了rbd存储

有一个问题，那就是开发人员他们并不是很了解yaml文件里面改怎么去写挂载，每种类型的存储都是不同的写法，那有没有一种方式屏蔽底层的写法，直接告诉k8s集群我想要一个什么样的存储呢？

有的，那就是pv

3. pv使用rbd

[root@master euler]# vim pvc.yaml 
apiVersion: v1
kind: PersistentVolumeClaim
metadata:
  name: myclaim
spec:
  accessModes:
    - ReadWriteOnce
  volumeMode: Block
  resources:
    requests:
      storage: 8Gi

这里的pvc使用的是块设备，8个G，目前还没有这个pv可以给到他

具体的这里不细说，CKA里面有写

注意，这里是pvc，并不是pv，pvc就是开发人员定义想要的存储类型，大小，然后我可以根据你的pvc去给你创建pv，或者提前创建好pv你直接申领

[root@master euler]# vim pv.yaml 
apiVersion: v1
kind: PersistentVolume
metadata:
  name: rbdpv
spec:
  capacity:
    storage: 8Gi
  volumeMode: Block
  accessModes:
    - ReadWriteOnce
  persistentVolumeReclaimPolicy: Recycle
  mountOptions:
    - hard
    - nfsvers=4.1
  rbd:
    fsType: xfs
    image: test
    keyring: /etc/ceph/ceph.client.admin.keyring
    monitors:
    - 172.16.1.33
    pool: rbd
    readOnly: false
    user: admin

3.1 查看pvc状态

[root@master euler]# kubectl get pvc
NAME      STATUS   VOLUME   CAPACITY   ACCESS MODES   STORAGECLASS   AGE
myclaim   Bound    rbdpv    8Gi        RWO                           11s

这个时候pv和就和pvc绑定上了，一个pv只能绑定一个pvc，同样，一个pvc也只能绑定一个pv

3.2 使用pvc

[root@master euler]# vim pod-pvc.yaml 
apiVersion: v1
kind: Pod
metadata:
  creationTimestamp: null
  labels:
    run: pvc-pod
  name: pvc-pod
spec:
  volumes:
  - name: rbd
    persistentVolumeClaim:
      claimName: myclaim
      readOnly: false
  containers:
  - image: nginx
    imagePullPolicy: IfNotPresent
    name: pvc-pod
    volumeDevices:             # 因为是使用的块设备，所以这里是volumeDevices
    - devicePath: /dev/rbd0
      name: rbd
    resources: {}
  dnsPolicy: ClusterFirst
  restartPolicy: Always
status: {}
~          

[root@master euler]# kubectl get pods
NAME                 READY   STATUS    RESTARTS   AGE
pvc-pod              1/1     Running   0          2m5s
rbd-5db4759c-v9cgm   1/1     Running   0          39m

3.3 进入容器查看块设备

root@pvc-pod:/# ls /dev/rbd0
/dev/rbd0

可以看到，现在rbd0已经存在于容器内部了

这样做我们每次创建pvc都需要创建对应的pv，我们可以使用动态制备

4. 动态制备

使用storageClass，但是目前欧拉使用的k8s太老了，所以需要下载欧拉fork的一个storageClass

[root@master ~]# git clone https://gitee.com/yftyxa/ceph-csi.git
[root@master ~]# cd ceph-csi/deploy/
[root@master deploy]# ls
ceph-conf.yaml  csi-config-map-sample.yaml  rbd
cephcsi         Makefile                    scc.yaml
cephfs          nfs                         service-monitor.yaml

4.1 动态制备rbd

我们需要修改/root/ceph-csi/deploy/rbd/kubernetes/csi-config-map.yaml

# 先创建一个csi命名空间
[root@master ~]# kubectl create ns csi

修改文件内容

[root@master kubernetes]# vim csi-rbdplugin-provisioner.yaml
# 将第63行的内容改为false
63             - "--extra-create-metadata=false"
# 修改第二个文件
[root@master kubernetes]# vim csi-config-map.yaml 
apiVersion: v1
kind: ConfigMap
metadata:
  name: "ceph-csi-config"
data:
  config.json: |-
    [
    {
      "clusterID": "c1f213ae-2de3-11ef-ae15-00163e179ce3",
      "monitors": ["172.16.1.33","172.16.1.32","172.16.1.31"]
    }
]

• 这里面的clusterID可以通过ceph -s去查看

修改第三个文件

[root@master kubernetes]# vim csidriver.yaml 
---
apiVersion: storage.k8s.io/v1
kind: CSIDriver
metadata:
  name: "rbd.csi.ceph.com"
spec:
  attachRequired: true
  podInfoOnMount: false
  # seLinuxMount: true       # 将这一行注释
  fsGroupPolicy: File

自行编写一个文件

[root@master kubernetes]# vim csi-kms-config-map.yaml 
---
apiVersion: v1
kind: ConfigMap
metadata:
  name: ceph-csi-encryption-kms-config
data:
  config-json: |-
    {}

4.2 获取admin的key

[root@ceph001 ~]# cat /etc/ceph/ceph.client.admin.keyring 
[client.admin]
	key = AQC4QnJmng4HIhAA42s27yOflqOBNtEWDgEmkg==
	caps mds = "allow *"
	caps mgr = "allow *"
	caps mon = "allow *"
	caps osd = "allow *"

• AQC4QnJmng4HIhAA42s27yOflqOBNtEWDgEmkg== 只要这部分

然后自行编写一个csi-secret.yaml的文件

[root@master kubernetes]# vim csi-secret.yaml
apiVersion: v1
kind: Secret
metadata:
  name: csi-secret
stringData:
  userID: admin
  userKey: AQC4QnJmng4HIhAA42s27yOflqOBNtEWDgEmkg==
  adminID: admin
  adminKey: AQC4QnJmng4HIhAA42s27yOflqOBNtEWDgEmkg==                     
[root@master kubernetes]# kubectl apply -f csi-secret.yaml -n csi
secret/csi-secret created
[root@master kubernetes]# cd ../../
[root@master deploy]# kubectl apply -f ceph-conf.yaml -n csi
configmap/ceph-config created
[root@master deploy]# cd -
/root/ceph-csi/deploy/rbd/kubernetes

4.3替换所有的namespace

[root@master kubernetes]# sed -i "s/namespace: default/namespace: csi/g" 
*.yaml

4.4 部署

[root@master kubernetes]# [root@master kubernetes]# kubectl apply -f . -n csi

注意：如果你的worker节点数量少于3个的话，是需要将 csi-rbdplugin-provisioner.yaml这个文件里面的replicas改小一点的。

[root@master kubernetes]# kubectl get pods -n csi
NAME                                         READY   STATUS    RESTARTS        AGE
csi-rbdplugin-cv455                          3/3     Running   1 (2m14s ago)   2m46s
csi-rbdplugin-pf5ld                          3/3     Running   0               4m36s
csi-rbdplugin-provisioner-6846c4df5f-dvqqk   7/7     Running   0               4m36s
csi-rbdplugin-provisioner-6846c4df5f-nmcxf   7/7     Running   1 (2m11s ago)   4m36s

5.使用动态制备

5.1 创建storageClass

[root@master rbd]# /root/ceph-csi/examples/rbd
[root@master rbd]# grep -Ev "\s*#|^$" storageclass.yaml 
---
apiVersion: storage.k8s.io/v1
kind: StorageClass
metadata:
   name: csi-rbd-sc
provisioner: rbd.csi.ceph.com
parameters:
   clusterID: <cluster-id>
   pool: <rbd-pool-name>
   imageFeatures: "layering"
   csi.storage.k8s.io/provisioner-secret-name: csi-rbd-secret
   csi.storage.k8s.io/provisioner-secret-namespace: default
   csi.storage.k8s.io/controller-expand-secret-name: csi-rbd-secret
   csi.storage.k8s.io/controller-expand-secret-namespace: default
   csi.storage.k8s.io/node-stage-secret-name: csi-rbd-secret
   csi.storage.k8s.io/node-stage-secret-namespace: default
   csi.storage.k8s.io/fstype: ext4
reclaimPolicy: Delete
allowVolumeExpansion: true
mountOptions:
   - discard

将这里的内容复制出来

---
apiVersion: storage.k8s.io/v1
kind: StorageClass
metadata:
   name: csi-rbd-sc
provisioner: rbd.csi.ceph.com
parameters:
   clusterID: c1f213ae-2de3-11ef-ae15-00163e179ce3
   pool: rbd
   imageFeatures: "layering"
   csi.storage.k8s.io/provisioner-secret-name: csi-secret
   csi.storage.k8s.io/provisioner-secret-namespace: csi
   csi.storage.k8s.io/controller-expand-secret-name: csi-secret
   csi.storage.k8s.io/controller-expand-secret-namespace: csi
   csi.storage.k8s.io/node-stage-secret-name: csi-secret
   csi.storage.k8s.io/node-stage-secret-namespace: csi
   csi.storage.k8s.io/fstype: ext4
reclaimPolicy: Retain
allowVolumeExpansion: true
mountOptions:
   - discard

修改成这个样子，这里面的clusterID改成自己的，secret-name自己查一下

5.2 创建pvc

[root@master euler]# cp pvc.yaml sc-pvc.yaml
[root@master euler]# vim sc-pvc.yaml
apiVersion: v1
kind: PersistentVolumeClaim
metadata:
  name: sc-pvc
spec:
  accessModes:
    - ReadWriteOnce
  volumeMode: Block
  storageClassName: "csi-rbd-sc"
  resources:
    requests:
      storage: 15Gi

• storageClassName 可以使用 kubectl get sc查看

现在我们只需要创建pvc，他就自己可以创建pv了

[root@master euler]# kubectl apply -f sc-pvc.yaml 
persistentvolumeclaim/sc-pvc created
[root@master euler]# kubectl get pvc
NAME      STATUS   VOLUME                                     CAPACITY   ACCESS MODES   STORAGECLASS   AGE
myclaim   Bound    rbdpv                                      8Gi        RWO                           111m
sc-pvc    Bound    pvc-dfe3497f-9ed7-4961-9265-9e7242073c28   15Gi       RWO            csi-rbd-sc     2s

回到ceph集群查看rbd

[root@ceph001 ~]# rbd ls
csi-vol-56e37046-b9d7-4ef1-a534-970a766744f3
test
[root@ceph001 ~]# rbd info csi-vol-56e37046-b9d7-4ef1-a534-970a766744f3
rbd image 'csi-vol-56e37046-b9d7-4ef1-a534-970a766744f3':
	size 15 GiB in 3840 objects
	order 22 (4 MiB objects)
	snapshot_count: 0
	id: 38019ee708da
	block_name_prefix: rbd_data.38019ee708da
	format: 2
	features: layering
	op_features: 
	flags: 
	create_timestamp: Wed Jun 19 04:55:35 2024
	access_timestamp: Wed Jun 19 04:55:35 2024
	modify_timestamp: Wed Jun 19 04:55:35 2024

5.3 将sc设为默认

如果不设置为默认的话，每次写yaml文件都需要指定sc,将sc设为默认的话就不用每次都指定了

[root@master euler]# kubectl edit sc csi-rbd-sc
# 在注释里面写入这一行
  annotations:
    storageclass.kubernetes.io/is-default-class: "true"

5.4 测试默认pvc

[root@master euler]# kubectl get sc
NAME                   PROVISIONER        RECLAIMPOLICY   VOLUMEBINDINGMODE   ALLOWVOLUMEEXPANSION   AGE
csi-rbd-sc (default)   rbd.csi.ceph.com   Retain          Immediate           true                   29m

再去查看sc就会有一个default的显示

[root@master euler]# cp sc-pvc.yaml sc-pvc1.yaml 
[root@master euler]# cat sc-pvc1.yaml 
apiVersion: v1
kind: PersistentVolumeClaim
metadata:
  name: sc-pvc1
spec:
  accessModes:
    - ReadWriteOnce
  volumeMode: Block
  resources:
    requests:
      storage: 20Gi

这个文件里面是没有指定storageClassName的

[root@master euler]# kubectl apply -f sc-pvc1.yaml 
persistentvolumeclaim/sc-pvc1 created
[root@master euler]# kubectl get pvc
NAME      STATUS   VOLUME                                     CAPACITY   ACCESS MODES   STORAGECLASS   AGE
myclaim   Bound    rbdpv                                      8Gi        RWO                           138m
sc-pvc    Bound    pvc-dfe3497f-9ed7-4961-9265-9e7242073c28   15Gi       RWO            csi-rbd-sc     27m
sc-pvc1   Bound    pvc-167cf73b-4983-4c28-aa98-bb65bb966649   20Gi       RWO            csi-rbd-sc     6s

这样就好了