一、有一个ceph cluster，假设已经准备好了，文档网上一大堆
二、开始集成ceph和kuberntes
2.1 禁用rbd features
rbd image有4个 features，layering, exclusive-lock, object-map, fast-diff, deep-flatten
因为目前内核仅支持layering，修改默认配置
每个ceph node的/etc/ceph/ceph.conf 添加一行
rbd_default_features = 1
这样之后创建的image 只有这一个feature
验证方式：
|
1
2
|
ceph --show-config|
grep
rbd|
grep
features
rbd_default_features = 1
|
2.2 创建ceph-secret这个k8s secret对象，这个secret对象用于k8s volume插件访问ceph集群：
获取client.admin的keyring值，并用base64编码：
|
1
2
|
# ceph auth get-key client.admin
AQBRIaFYqWT8AhAAUtmJgeNFW
/o1ylUzssQQhA
==
|
|
1
2
|
# echo "AQBRIaFYqWT8AhAAUtmJgeNFW/o1ylUzssQQhA=="|base64
QVFCUklhRllxV1Q4QWhBQVV0bUpnZU5GVy9vMXlsVXpzc1FRaEE9PQo=
|
创建ceph-secret.yaml文件，data下的key字段值即为上面得到的编码值：
|
1
2
3
4
5
6
|
apiVersion: v1
kind: Secret
metadata:
name: ceph-secret
data:
key: QVFCUklhRllxV1Q4QWhBQVV0bUpnZU5GVy9vMXlsVXpzc1FRaEE9PQo=
|
创建ceph-secret:
|
1
2
3
4
5
|
# kubectl create -f ceph-secret.yamlsecret "ceph-secret" created
# kubectl get secret
NAME TYPE DATA AGE
ceph-secret Opaque 1 2d
default-token-5vt3n kubernetes.io
/service-account-token
3 106d
|
三、Kubernetes Persistent Volume和Persistent Volume Claim
概念：PV是集群的资源，PVC请求资源并检查资源是否可用
注意：以下操作设计到name的参数，一定要一致
3.1 创建disk image (以jdk保存到ceph举例)
|
1
2
3
4
5
6
7
8
9
|
# rbd create jdk-image -s 1G
# rbd info jdk-image
rbd image
'jdk-image'
:
size 1024 MB
in
256 objects
order 22 (4096 kB objects)
block_name_prefix: rbd_data.37642ae8944a
format
: 2
features: layering
flags:
|
3.2 创建pv(仍然使用之前创建的ceph-secret)
创建jdk-pv.yaml:
monitors: 就是ceph的mon，有几个写几个
|
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
|
apiVersion: v1
kind: PersistentVolume
metadata:
name: jdk-pv
spec:
capacity:
storage: 2Gi
accessModes:
- ReadWriteOnce
rbd:
monitors:
- 10.10.10.1:6789
pool: rbd
image: jdk-image
user: admin
secretRef:
name: ceph-secret
fsType: xfs
readOnly:
false
persistentVolumeReclaimPolicy: Recycle
|
执行创建操作：
|
1
2
3
4
5
|
# kubectl create -f jdk-pv.yamlpersistentvolume "jdk-pv" created
#kubectl get pv
NAME CAPACITY ACCESSMODES RECLAIMPOLICY STATUS CLAIM REASON AGE
ceph-pv 1Gi RWO Recycle Bound default
/ceph-claim
1d
jdk-pv 2Gi RWO Recycle Available 1m
|
3.3 创建pvc
创建jdk-pvc.yaml
|
1
2
3
4
5
6
7
8
9
10
|
kind: PersistentVolumeClaim
apiVersion: v1
metadata:
name: jdk-claim
spec:
accessModes:
- ReadWriteOnce
resources:
requests:
storage: 1Gi
|
执行创建操作：
|
1
2
3
4
5
|
# kubectl create -f jdk-pvc.yamlpersistentvolumeclaim "jdk-claim" created
# kubectl get pvc
NAME STATUS VOLUME CAPACITY ACCESSMODES AGE
ceph-claim Bound ceph-pv 1Gi RWO 2d
jdk-claim Bound jdk-pv 2Gi RWO 39s
|
3.4 创建挂载ceph rbd的pod:
创建 ceph-busyboxpod.yaml
|
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
|
apiVersion: v1
kind: Pod
metadata:
name: ceph-busybox
spec:
containers:
- name: ceph-busybox
image: busybox
command
: [
"sleep"
,
"600000"
]
volumeMounts:
- name: ceph-vol1
mountPath:
/usr/share/busybox
readOnly:
false
volumes:
- name: ceph-vol1
persistentVolumeClaim:
claimName: jdk-claim
|
执行创建操作：
|
1
|
kubectl create -f ceph-busyboxpod.yaml
|
ceph rbd 持久化 这里描述下：
1、稳定性在于ceph
2、只能同一node挂载，不能跨node
3、读写只能一个pod，其他pod只能读
官方url描述
https://kubernetes.io/docs/user-guide/volumes/#rbd
附官方关于kubernetes的volume的mode
https://kubernetes.io/docs/user-guide/persistent-volumes/
本文转自银狐博客51CTO博客，原文链接http://blog.51cto.com/foxhound/1899545如需转载请自行联系原作者
战狐
