參考文檔
https://github.com/kubernetes/examples/tree/master/staging/volumes/rbd
http://docs.ceph.com/docs/mimic/rados/operations/pools/
https://blog.csdn.net/aixiaoyang168/article/details/78999851
https://www.cnblogs.com/keithtt/p/6410302.html
https://kubernetes.io/docs/concepts/storage/volumes/
https://kubernetes.io/docs/concepts/storage/persistent-volumes/
https://blog.csdn.net/wenwenxiong/article/details/78406136
http://www.mamicode.com/info-detail-1701743.html
簡介
ceph支持對象存儲，文件系統及塊存儲，是三合一存儲類型，kubernetes的樣例中有cephfs與rbd兩種使用方式的介紹，cephfs需要node節點安裝ceph才能支持，rbd需要node節點安裝ceph-common才支持。
使用上的區別如下：
Volume Plugin ReadWriteOnce ReadOnlyMany ReadWriteMany
CephFS ✓ ✓ ✓
RBD ✓ ✓ -
基本環境
k81集羣1.13.1版本
[root@elasticsearch01 ~]# kubectl get nodes
NAME STATUS ROLES AGE VERSION
10.2.8.34 Ready <none> 24d v1.13.1
10.2.8.65 Ready <none> 24d v1.13.1
ceph集羣 luminous版本
[root@ceph01 ~]# ceph -s
services:
mon: 3 daemons, quorum ceph01,ceph02,ceph03
mgr: ceph03(active), standbys: ceph02, ceph01
osd: 24 osds: 24 up, 24 in
rgw: 3 daemons active
操作步驟
一、ceph集羣創建ceph池及鏡像
[root@ceph01 ~]# ceph osd pool create rbd-k8s 1024 1024
For better initial performance on pools expected to store a large number of objects, consider supplying the expected_num_objects parameter when creating the pool.
[root@ceph01 ~]# ceph osd lspools
1 rbd-es,2 .rgw.root,3 default.rgw.control,4 default.rgw.meta,5 default.rgw.log,6 default.rgw.buckets.index,7 default.rgw.buckets.data,8 default.rgw.buckets.non-ec,9 rbd-k8s,
[root@ceph01 ~]# rbd create rbd-k8s/cephimage1 --size 10240
[root@ceph01 ~]# rbd create rbd-k8s/cephimage2 --size 20480
[root@ceph01 ~]# rbd create rbd-k8s/cephimage3 --size 40960
[root@ceph01 ~]# rbd list rbd-k8s
cephimage1
cephimage2
cephimage3
二、k8s集羣使用ceph rbd塊存儲
1、下載樣例
[root@elasticsearch01 ~]# git clone https://github.com/kubernetes/examples.git
Cloning into 'examples'...
remote: Enumerating objects: 11475, done.
remote: Total 11475 (delta 0), reused 0 (delta 0), pack-reused 11475
Receiving objects: 100% (11475/11475), 16.94 MiB | 6.00 MiB/s, done.
Resolving deltas: 100% (6122/6122), done.
[root@elasticsearch01 ~]# cd examples/staging/volumes/rbd
[root@elasticsearch01 rbd]# ls
rbd-with-secret.yaml rbd.yaml README.md secret
[root@elasticsearch01 rbd]# cp -a ./rbd /k8s/yaml/volumes/
2、k8s集羣節點安裝ceph客戶端
[root@elasticsearch01 ceph]# yum install ceph-common
3、修改rbd-with-secret.yaml配置文件
修改後配置如下：
[root@elasticsearch01 rbd]# cat rbd-with-secret.yaml
apiVersion: v1
kind: Pod
metadata:
name: rbd2
spec:
containers:
- image: kubernetes/pause
name: rbd-rw
volumeMounts:
- name: rbdpd
mountPath: /mnt/rbd
volumes:
- name: rbdpd
rbd:
monitors:
- '10.0.4.10:6789'
- '10.0.4.13:6789'
- '10.0.4.15:6789'
pool: rbd-k8s
image: cephimage1
fsType: ext4
readOnly: true
user: admin
secretRef:
name: ceph-secret
如下參數根據實際情況修改：
monitors：這是 Ceph集羣的monitor 監視器，Ceph 集羣可以配置多個 monitor，本配置3個mon
pool：這是Ceph集羣中存儲數據進行歸類區分使用，這裏用的pool爲rbd-ceph
image：這是Ceph 塊設備中的磁盤映像文件，這裏用的是cephimage1
fsType：文件系統類型，默認使用 ext4 即可
readOnly：是否爲只讀，這裏測試使用只讀即可
user：這是Ceph Client訪問Ceph存儲集羣所使用的用戶名，這裏我們使用admin 即可
keyring：這是Ceph集羣認證需要的密鑰環，搭建Ceph存儲集羣時生成的ceph.client.admin.keyring
imageformat：這是磁盤映像文件格式，可以使用 2，或者老一些的1，內核版本比較低的使用1
imagefeatures： 這是磁盤映像文件的特徵，需要uname -r查看集羣系統內核所支持的特性，這裏Ceontos7.4內核版本爲3.10.0-693.el7.x86_64只支持layering
4、使用ceph認證祕鑰
在集羣中使用secret更方便易於擴展且安全
[root@ceph01 ~]# cat /etc/ceph/ceph.client.admin.keyring
[client.admin]
key = AQBHVp9bPirBCRAAUt6Mjw5PUjiy/RDHyHZrUw==
[root@ceph01 ~]# grep key /etc/ceph/ceph.client.admin.keyring |awk '{printf "%s", $NF}'|base64
QVFCSFZwOWJQaXJCQ1JBQVV0Nk1qdzVQVWppeS9SREh5SFpyVXc9PQ==
5、創建ceph-secret
[root@elasticsearch01 rbd]# cat secret/ceph-secret.yaml
apiVersion: v1
kind: Secret
metadata:
name: ceph-secret
type: "kubernetes.io/rbd"
data:
key: QVFCSFZwOWJQaXJCQ1JBQVV0Nk1qdzVQVWppeS9SREh5SFpyVXc9PQ==
[root@elasticsearch01 rbd]# kubectl create -f secret/ceph-secret.yaml
secret/ceph-secret created
6、創建pod測試rbd
按照官網的案例直接創建即可
[root@elasticsearch01 rbd]# kubectl create -frbd-with-secret.yaml
但是生產環境中不直接使用volumes，他會隨着pods的創建兒創建，刪除而刪除，數據得不到保存，如果需要數據不丟失，需要藉助pv和pvc實現
7、創建ceph-pv
注意rbd是讀寫一次，只讀多次，目前還不支持讀寫多次，我們日常使用rbd映射磁盤時也是一個image只掛載一個客戶端上；cephfs可以支持讀寫多次
[root@elasticsearch01 rbd]# cat rbd-pv.yaml
apiVersion: v1
kind: PersistentVolume
metadata:
name: ceph-rbd-pv
spec:
capacity:
storage: 20Gi
accessModes:
- ReadWriteOnce
rbd:
monitors:
- '10.0.4.10:6789'
- '10.0.4.13:6789'
- '10.0.4.15:6789'
pool: rbd-k8s
image: cephimage2
user: admin
secretRef:
name: ceph-secret
fsType: ext4
readOnly: false
persistentVolumeReclaimPolicy: Recycle
[root@elasticsearch01 rbd]# kubectl create -f rbd-pv.yaml
persistentvolume/ceph-rbd-pv created
[root@elasticsearch01 rbd]# kubectl get pv
NAME CAPACITY ACCESS MODES RECLAIM POLICY STATUS CLAIM STORAGECLASS REASON AGE
ceph-rbd-pv 20Gi RWO Recycle Available
8、創建ceph-pvc
[root@elasticsearch01 rbd]# cat rbd-pv-claim.yaml
apiVersion: v1
kind: PersistentVolumeClaim
metadata:
name: ceph-rbd-pv-claim
spec:
accessModes:
- ReadWriteOnce
resources:
requests:
storage: 10Gi
[root@elasticsearch01 rbd]# kubectl create -f rbd-pv-claim.yaml
persistentvolumeclaim/ceph-rbd-pv-claim created
[root@elasticsearch01 rbd]# kubectl get pvc
NAME STATUS VOLUME CAPACITY ACCESS MODES STORAGECLASS AGE
ceph-rbd-pv-claim Bound ceph-rbd-pv 20Gi RWO 6s
[root@elasticsearch01 rbd]# kubectl get pv
NAME CAPACITY ACCESS MODES RECLAIM POLICY STATUS CLAIM STORAGECLASS REASON AGE
ceph-rbd-pv 20Gi RWO Recycle Bound default/ceph-rbd-pv-claim 5m28s
9、創建pod通過pv、pvc方式測試rbd
由於需要格式化掛載rbd，rbd空間比較大10G，需要時間比較久，大概需要幾分鐘
[root@elasticsearch01 rbd]# cat rbd-pv-pod.yaml
apiVersion: v1
kind: Pod
metadata:
name: ceph-rbd-pv-pod1
spec:
containers:
- name: ceph-rbd-pv-busybox
image: busybox
command: ["sleep", "60000"]
volumeMounts:
- name: ceph-rbd-vol1
mountPath: /mnt/ceph-rbd-pvc/busybox
readOnly: false
volumes:
- name: ceph-rbd-vol1
persistentVolumeClaim:
claimName: ceph-rbd-pv-claim
[root@elasticsearch01 rbd]# kubectl create -f rbd-pv-pod.yaml
pod/ceph-rbd-pv-pod1 created
[root@elasticsearch01 rbd]# kubectl get pods
NAME READY STATUS RESTARTS AGE
busybox 1/1 Running 432 18d
ceph-rbd-pv-pod1 0/1 ContainerCreating 0 19s
報錯如下
MountVolume.WaitForAttach failed for volume "ceph-rbd-pv" : rbd: map failed exit status 6, rbd output: rbd: sysfs write failed RBD image feature set mismatch. Try disabling features unsupported by the kernel with "rbd feature disable". In some cases useful info is found in syslog - try "dmesg | tail". rbd: map failed: (6) No such device or address
解決方法
禁用一些特性，這些特性在centos7.4內核上不支持，所以生產環境中k8s及相關ceph最好使用內核版本高的系統做爲底層操作系統
rbd feature disable rbd-k8s/cephimage2 exclusive-lock object-map fast-diff deep-flatten
[root@ceph01 ~]# rbd feature disable rbd-k8s/cephimage2 exclusive-lock object-map fast-diff deep-flatten
三、驗證效果
1、k8s集羣端驗證
[root@elasticsearch01 rbd]# kubectl get pods -o wide
NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES
busybox 1/1 Running 432 18d 10.254.35.3 10.2.8.65 <none> <none>
ceph-rbd-pv-pod1 1/1 Running 0 3m39s 10.254.35.8 10.2.8.65 <none> <none>
[root@elasticsearch02 ceph]# df -h |grep rbd
/dev/rbd0 493G 162G 306G 35% /data
/dev/rbd1 20G 45M 20G 1% /var/lib/kubelet/plugins/kubernetes.io/rbd/mounts/rbd-k8s-image-cephimage2
[root@elasticsearch01 rbd]# kubectl exec -ti ceph-rbd-pv-pod1 sh
/ # df -h
Filesystem Size Used Available Use% Mounted on
overlay 49.1G 7.4G 39.1G 16% /
tmpfs 64.0M 0 64.0M 0% /dev
tmpfs 7.8G 0 7.8G 0% /sys/fs/cgroup
/dev/vda1 49.1G 7.4G 39.1G 16% /dev/termination-log
/dev/vda1 49.1G 7.4G 39.1G 16% /etc/resolv.conf
/dev/vda1 49.1G 7.4G 39.1G 16% /etc/hostname
/dev/vda1 49.1G 7.4G 39.1G 16% /etc/hosts
shm 64.0M 0 64.0M 0% /dev/shm
/dev/rbd1 19.6G 44.0M 19.5G 0% /mnt/ceph-rbd-pvc/busybox
tmpfs 7.8G 12.0K 7.8G 0% /var/run/secrets/kubernetes.io/serviceaccount
tmpfs 7.8G 0 7.8G 0% /proc/acpi
tmpfs 64.0M 0 64.0M 0% /proc/kcore
tmpfs 64.0M 0 64.0M 0% /proc/keys
tmpfs 64.0M 0 64.0M 0% /proc/timer_list
tmpfs 64.0M 0 64.0M 0% /proc/timer_stats
tmpfs 64.0M 0 64.0M 0% /proc/sched_debug
tmpfs 7.8G 0 7.8G 0% /proc/scsi
tmpfs 7.8G 0 7.8G 0% /sys/firmware
/ # cd /mnt/ceph-rbd-pvc/busybox/
/mnt/ceph-rbd-pvc/busybox # ls
lost+found
/mnt/ceph-rbd-pvc/busybox # touch ceph-rbd-pods
/mnt/ceph-rbd-pvc/busybox # ls
ceph-rbd-pods lost+found
/mnt/ceph-rbd-pvc/busybox # echo busbox>ceph-rbd-pods
/mnt/ceph-rbd-pvc/busybox # cat ceph-rbd-pods
busbox
2、ceph集羣端驗證
[root@ceph01 ~]# ceph df
GLOBAL:
SIZE AVAIL RAW USED %RAW USED
65.9TiB 58.3TiB 7.53TiB 11.43
POOLS:
NAME ID USED %USED MAX AVAIL OBJECTS
rbd-es 1 1.38TiB 7.08 18.1TiB 362911
.rgw.root 2 1.14KiB 0 18.1TiB 4
default.rgw.control 3 0B 0 18.1TiB 8
default.rgw.meta 4 46.9KiB 0 104GiB 157
default.rgw.log 5 0B 0 18.1TiB 345
default.rgw.buckets.index 6 0B 0 104GiB 2012
default.rgw.buckets.data 7 1.01TiB 5.30 18.1TiB 2090721
default.rgw.buckets.non-ec 8 0B 0 18.1TiB 0
rbd-k8s 9 137MiB 0 18.1TiB 67