1)Kubernetes无状态服务
Kubernetes无状态服务特征:
1)是指该服务运行的实例不会在本地存储需要持久化的数据,并且多个实例对于同一请求响应的结果是完全一致的;
2)多个实例可以共享相同的持久化数据。例如:nginx实例、tomcat实例等;
3)相关的Kubernetes资源有:ReplicaSet、ReplicationController、Deployment等,由于是无状态服务,所以这些控制器创建的Pod名称都是随机性的。并且在缩容时并不会明确缩容某一个Pod,而是随机的,因为所有实例得到的返回值都是一样的,所以缩容任何一个Pod都可以;
2)Kubernetes有状态服务
Kubernetes有状态服务特征:
1)有状态服务可以说是需要数据存储功能的服务、或者指多线程类型的服务、队列等。(比如:mysql数据库、kafka、zookeeper等);
2)每个实例都需要自己独立的持久化存储,并且在Kubernetes中通过声明模板的方式来进行定义。持久卷声明模板在创建pod之前创建,绑定到pod中,模板可以定义多个;
3)相关的Kubernetes资源有:StatefulSet。由于是有状态的服务,所以每个Pod都有特定的名称和网络标识。比如Pod名称是由StatefulSet名+有序的数字组成(0、1、2……);
4)在进行缩容操作时,可以明确知道会缩容那一个Pod,从数字最大的开始。并且StatefulSet在已有实例不健康的情况下是不允许做缩容操作的;
3)无状态服务和有状态服务的区别
主要表现在以下方面:
1)实例数量:无状态服务可以有一个或多个实例,因此支持两种服务容量调节模式;有状态服务职能有一个实例,不允许创建多个实例,因此也不支持服务容量的调节;
2)存储卷:无状态服务可以有存储卷,也可以没有,即使有也无法备份存储卷中的数据;有状态服务必须要有存储卷,并且在创建服务时,必须指定该存储卷分配的磁盘空间大小;
3)数据存储:
无状态服务运行过程中的所有数据(除日志和监控数据)都存在容器实例里的文件系统中,如果实例停止或者删除,则这些数据都将丢失,无法找回;而对于有状态服务,凡是已经挂载了存储卷的目录下的文件内容都可以随时进行备份,备份的数据可以下载,也可以用于恢复新的服务。但对于没有挂载卷的目录下的数据,仍然是无法备份和保存的,如果实例停止或者删除,这些非挂载卷里的文件内容同样会丢失。
4)StatefulSet概述
StatefulSet是Kubernetes提供的管理有状态应用的负载管理控制器API。在Pods管理的基础上,保证Pods的顺序和一致性。与Deployment一样,StatefulSet也是使用容器的Spec来创建Pod,与之不同StatefulSet创建的Pods在生命周期中会保持持久的标记(例如Pod Name)。
5)StatefulSet特点
1)稳定的持久化存储,即Pod重新调度后还是能访问到相同的持久化数据,基于PVC来实现;
2)稳定的网络标志,即Pod重新调度后其PodName和HostName不变,基于Headless Service(即没有Cluster IP的Service)来实现;
3)有序部署,有序扩展,即Pod是有顺序的,在部署或者扩展的时候要依据定义的顺序依次依次进行(即从0到N-1,在下一个Pod运行之前所有之前的Pod必须都是Running和Ready状态),基于init containers来实现;
4)有序收缩,有序删除(即从N-1到0);
1)搭建NFS共享存储
为了方便,就直接在master节点上部署NFS存储了!
[root@master ~]# yum -y install nfs-utils rpcbind
[root@master ~]# vim /etc/exports
/nfsdata *(rw,sync,no_root_squash)
[root@master ~]# mkdir /nfsdata
[root@master ~]# systemctl start nfs-server
[root@master ~]# systemctl start rpcbind
[root@master ~]# showmount -e
Export list for master:
/nfsdata *
2)创建rbac授权
[root@master ~]# vim rbac-rolebind.yaml
apiVersion: v1 #创建一个用于认证的服务账号
kind: ServiceAccount
metadata:
name: nfs-provisioner
---
apiVersion: rbac.authorization.k8s.io/v1 #创建群集规则
kind: ClusterRole
metadata:
name: nfs-provisioner-runner
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: ["watch", "create", "update", "patch"]
- apiGroups: [""]
resources: ["services", "endpoints"]
verbs: ["get","create","list", "watch","update"]
- apiGroups: ["extensions"]
resources: ["podsecuritypolicies"]
resourceNames: ["nfs-provisioner"]
verbs: ["use"]
---
kind: ClusterRoleBinding #将服务认证用户与群集规则进行绑定
apiVersion: rbac.authorization.k8s.io/v1
metadata:
name: run-nfs-provisioner
subjects:
- kind: ServiceAccount
name: nfs-provisioner
namespace: default #必写字段,否则会提示错误
roleRef:
kind: ClusterRole
name: nfs-provisioner-runner
apiGroup: rbac.authorization.k8s.io
[root@master ~]# kubectl apply -f rbac-rolebind.yaml
3)创建nfs-deployment.资源
[root@master ~]# vim nfs-deployment.yaml
apiVersion: extensions/v1beta1
kind: Deployment
metadata:
name: nfs-client-provisioner
spec:
replicas: 1 #指定副本数量为1
strategy:
type: Recreate #指定策略类型为重置
template:
metadata:
labels:
app: nfs-client-provisioner
spec:
serviceAccount: nfs-provisioner #指定rbac yanl文件中创建的认证用户账号
containers:
- name: nfs-client-provisioner
image: registry.cn-hangzhou.aliyuncs.com/open-ali/nfs-client-provisioner #使用的镜像
volumeMounts:
- name: nfs-client-root
mountPath: /persistentvolumes #指定容器内挂载的目录
env:
- name: PROVISIONER_NAME #容器内的变量用于指定提供存储的名称
value: zjz
- name: NFS_SERVER #容器内的变量用于指定nfs服务的IP地址
value: 192.168.10.52
- name: NFS_PATH #容器内的变量指定nfs服务器对应的目录
value: /nfsdata
volumes: #指定挂载到容器内的nfs的路径及IP
- name: nfs-client-root
nfs:
server: 192.168.10.52
path: /nfsdata
[root@master ~]# kubectl apply -f nfs-deployment.yaml
[root@master ~]# kubectl get pod
NAME READY STATUS RESTARTS AGE
nfs-client-provisioner-74cf5dd55f-7fvr5 1/1 Running 0 44s
4)创建SC(Storage Class)
[root@master ~]# vim sc.yaml
apiVersion: storage.k8s.io/v1
kind: StorageClass
metadata:
name: stateful-nfs
namespace: xiaojiang-test
provisioner: zjz #这个要和nfs-client-provisioner的env环境变量中的PROVISIONER_NAME的value值对应。
reclaimPolicy: Retain #指定回收策略为Retain(手动释放)
[root@master ~]# kubectl apply -f sc.yaml
[root@master ~]# kubectl get StorageClass
NAME PROVISIONER AGE
stateful-nfs zjz 7s
5)创建Pod
[root@master ~]# vim statefulset.yaml
apiVersion: v1
kind: Service
metadata:
name: headless-svc #从名称就可以是无头服务
labels:
app: headless-svc
spec:
ports:
- port: 80
name: myweb
selector:
app: headless-pod
clusterIP: None #不分配群集的IP地址,所以不具备负载均衡的能力
---
apiVersion: apps/v1
kind: StatefulSet #定义pod中运行的应用
metadata:
name: statefulset-test
spec:
serviceName: headless-svc
replicas: 3
selector:
matchLabels:
app: headless-pod
template:
metadata:
labels:
app: headless-pod
spec:
containers:
- image: httpd
name: myhttpd
ports:
- containerPort: 80
name: httpd
volumeMounts:
- mountPath: /usr/local/apache2/htdocs
name: test
volumeClaimTemplates: #定义创建PVC使用的模板
- metadata:
name: test
annotations: #这是指定storageclass
volume.beta.kubernetes.io/storage-class: stateful-nfs
spec:
accessModes:
- ReadWriteOnce
resources:
requests:
storage: 100Mi
[root@master ~]# kubectl apply -f statefulset.yaml
[root@master ~]# kubectl get pod #这里需要等待一会儿
NAME READY STATUS RESTARTS AGE
nfs-client-provisioner-74cf5dd55f-7fvr5 1/1 Running 0 4m54s
statefulset-test-0 1/1 Running 0 94s
statefulset-test-1 1/1 Running 0 55s
statefulset-test-2 1/1 Running 0 20s
[root@master ~]# kubectl get pv #查看pv
NAME CAPACITY ACCESS MODES RECLAIM POLICY STATUS CLAIM STORAGECLASS REASON AGE
pvc-6c4d3c6f-6279-4ea6-ad8e-10d3098ea0eb 100Mi RWO Delete Bound default/test-statefulset-test-2 stateful-nfs 61s
pvc-d01c09ab-1399-49d9-bde6-812ad557f952 100Mi RWO Delete Bound default/test-statefulset-test-0 stateful-nfs 2m15s
pvc-e2257b8f-9deb-4398-85ed-b540a5880e08 100Mi RWO Delete Bound default/test-statefulset-test-1 stateful-nfs 96s
[root@master ~]# kubectl get pvc #查看pvc
NAME STATUS VOLUME CAPACITY ACCESS MODES STORAGECLASS AGE
test-statefulset-test-0 Bound pvc-d01c09ab-1399-49d9-bde6-812ad557f952 100Mi RWO stateful-nfs 2m20s
test-statefulset-test-1 Bound pvc-e2257b8f-9deb-4398-85ed-b540a5880e08 100Mi RWO stateful-nfs 101s
test-statefulset-test-2 Bound pvc-6c4d3c6f-6279-4ea6-ad8e-10d3098ea0eb 100Mi RWO stateful-nfs 66s
[root@master ~]# ls /nfsdata/
default-test-statefulset-test-0-pvc-d01c09ab-1399-49d9-bde6-812ad557f952
default-test-statefulset-test-1-pvc-e2257b8f-9deb-4398-85ed-b540a5880e08
default-test-statefulset-test-2-pvc-6c4d3c6f-6279-4ea6-ad8e-10d3098ea0eb
#pod里写测试内容
[root@master ~]# echo "0000" > /nfsdata/default-test-statefulset-test-0-pvc-d01c09ab-1399-49d9-bde6-812ad557f952/index.html
[root@master ~]# echo "1111" > /nfsdata/default-test-statefulset-test-1-pvc-e2257b8f-9deb-4398-85ed-b540a5880e08/index.html
[root@master ~]# echo "2222" > /nfsdata/default-test-statefulset-test-2-pvc-6c4d3c6f-6279-4ea6-ad8e-10d3098ea0eb/index.html
#查看podip进行测试
[root@master ~]# kubectl get pod -o wide
NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES
nfs-client-provisioner-74cf5dd55f-7fvr5 1/1 Running 0 8m49s 10.244.1.2 node01
statefulset-test-0 1/1 Running 0 5m29s 10.244.2.2 node02
statefulset-test-1 1/1 Running 0 4m50s 10.244.1.3 node01
statefulset-test-2 1/1 Running 0 4m15s 10.244.2.3 node02
#测试
[root@master ~]# curl 10.244.1.3
1111
[root@master ~]# curl 10.244.2.3
2222
[root@master ~]# curl 10.244.2.2
0000
[root@master ~]# curl -I 10.244.2.2
HTTP/1.1 200 OK
Date: Fri, 21 Aug 2020 03:45:16 GMT
Server: Apache/2.4.46 (Unix)
Last-Modified: Fri, 21 Aug 2020 03:33:50 GMT
ETag: "5-5ad5ae7c9eb7a"
Accept-Ranges: bytes
Content-Length: 5
Content-Type: text/html
#可以看出现在提供web页面的服务是Apache
#在pod里相应的目录下查看建立的测试内容
[root@master ~]# kubectl exec statefulset-test-0 cat /usr/local/apache2/htdocs/index.html
0000
[root@master ~]# kubectl exec statefulset-test-1 cat /usr/local/apache2/htdocs/index.html
1111
[root@master ~]# kubectl exec statefulset-test-2 cat /usr/local/apache2/htdocs/index.html
2222
#删除一个pod,测试数据是否会丢失
[root@master ~]# kubectl delete pod statefulset-test-2
pod "statefulset-test-2" deleted
#删除之后肯定会建立新的pod
[root@master ~]# kubectl get pod
NAME READY STATUS RESTARTS AGE
nfs-client-provisioner-74cf5dd55f-7fvr5 1/1 Running 0 15m
statefulset-test-0 1/1 Running 0 12m
statefulset-test-1 1/1 Running 0 11m
statefulset-test-2 0/1 ContainerCreating 0 12s
#建立新的pod完成
[root@master ~]# kubectl get pod
NAME READY STATUS RESTARTS AGE
nfs-client-provisioner-74cf5dd55f-7fvr5 1/1 Running 0 15m
statefulset-test-0 1/1 Running 0 12m
statefulset-test-1 1/1 Running 0 11m
statefulset-test-2 1/1 Running 0 18s
#再次查看数据
[root@master ~]# kubectl exec statefulset-test-2 cat /usr/local/apache2/htdocs/index.html
2222
6)对pod进行更新并扩容 (注释#的地方注意修改)
[root@master ~]# vim statefulset.yaml
apiVersion: v1
kind: Service
metadata:
name: headless-svc
labels:
app: headless-svc
spec:
ports:
- port: 80
name: myweb
selector:
app: headless-pod
clusterIP: None
---
apiVersion: apps/v1
kind: StatefulSet
metadata:
name: statefulset-test
spec:
updateStrategy:
rollingUpdate:
partition: 2 #默认值为0(表示所有都会更新),值为2表示第三个pod进行更新
serviceName: headless-svc
replicas: 10
selector:
matchLabels:
app: headless-pod
template:
metadata:
labels:
app: headless-pod
spec:
containers:
- image: nginx #更换扩容时使用的镜像
name: myhttpd
ports:
- containerPort: 80
name: httpd
volumeMounts:
- mountPath: /usr/share/nginx/html/ #更换容器中的主目录
name: test
volumeClaimTemplates:
- metadata:
name: test
annotations: #这是指定storageclass
volume.beta.kubernetes.io/storage-class: stateful-nfs
spec:
accessModes:
- ReadWriteOnce
resources:
requests:
storage: 100Mi
[root@master ~]# kubectl apply -f statefulset.yaml
[root@master ~]# kubectl get pod -o wide #需要等待一会儿
NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES
nfs-client-provisioner-74cf5dd55f-7fvr5 1/1 Running 0 31m 10.244.1.2 node01
statefulset-test-0 1/1 Running 0 28m 10.244.2.2 node02
statefulset-test-1 1/1 Running 0 28m 10.244.1.3 node01
statefulset-test-2 0/1 Terminating 0 16m 10.244.2.4 node02
statefulset-test-3 1/1 Running 0 2m34s 10.244.1.4 node01
statefulset-test-4 1/1 Running 0 2m10s 10.244.2.5 node02
statefulset-test-5 1/1 Running 0 102s 10.244.1.5 node01
statefulset-test-6 1/1 Running 0 83s 10.244.1.6 node01
statefulset-test-7 1/1 Running 0 63s 10.244.2.6 node02
statefulset-test-8 1/1 Running 0 42s 10.244.1.7 node01
statefulset-test-9 1/1 Running 0 23s 10.244.2.7 node02
[root@master ~]# ls /nfsdata/
default-test-statefulset-test-0-pvc-d01c09ab-1399-49d9-bde6-812ad557f952
default-test-statefulset-test-1-pvc-e2257b8f-9deb-4398-85ed-b540a5880e08
default-test-statefulset-test-2-pvc-6c4d3c6f-6279-4ea6-ad8e-10d3098ea0eb
default-test-statefulset-test-3-pvc-95724e2b-e30a-4c8c-a56d-9b9aa46d286c
default-test-statefulset-test-4-pvc-8d93522b-ccf6-492e-9adf-93a933214398
default-test-statefulset-test-5-pvc-cfbe5307-e756-487e-b03f-2edc62a86884
default-test-statefulset-test-6-pvc-e965f491-e04e-4488-9ee9-8069b2a224f1
default-test-statefulset-test-7-pvc-b9ca6c86-63f1-4b3a-9461-d1e844029d97
default-test-statefulset-test-8-pvc-7ff2c99a-34fe-4de3-bb70-d842d68b74d7
default-test-statefulset-test-9-pvc-3f01625d-28ba-4655-a3c8-54a4b8a89a81
[root@master ~]# ls /nfsdata/ | wc -l
10
[root@master ~]# kubectl get pod -o wide
NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES
nfs-client-provisioner-74cf5dd55f-7fvr5 1/1 Running 0 33m 10.244.1.2 node01
statefulset-test-0 1/1 Running 0 29m 10.244.2.2 node02
statefulset-test-1 1/1 Running 0 29m 10.244.1.3 node01
statefulset-test-2 1/1 Running 0 62s 10.244.2.8 node02
statefulset-test-3 1/1 Running 0 3m45s 10.244.1.4 node01
statefulset-test-4 1/1 Running 0 3m21s 10.244.2.5 node02
statefulset-test-5 1/1 Running 0 2m53s 10.244.1.5 node01
statefulset-test-6 1/1 Running 0 2m34s 10.244.1.6 node01
statefulset-test-7 1/1 Running 0 2m14s 10.244.2.6 node02
statefulset-test-8 1/1 Running 0 113s 10.244.1.7 node01
statefulset-test-9 1/1 Running 0 94s 10.244.2.7 node02
#由于配置文件中写的是 partition: 2 ,(2表示第三个pod进行更新),所以statefulset-test-0 statefulset-test-1 statefulset-test-2是apache,后面的是nginx
[root@master ~]# curl -I 10.244.2.2
HTTP/1.1 200 OK
Date: Fri, 21 Aug 2020 04:02:20 GMT
Server: Apache/2.4.46 (Unix)
Last-Modified: Fri, 21 Aug 2020 03:33:50 GMT
ETag: "5-5ad5ae7c9eb7a"
Accept-Ranges: bytes
Content-Length: 5
Content-Type: text/html
[root@master ~]# curl -I 10.244.1.4
HTTP/1.1 403 Forbidden
Server: nginx/1.19.2
Date: Fri, 21 Aug 2020 04:02:37 GMT
Content-Type: text/html
Content-Length: 153
Connection: keep-alive
由此可以看出在在扩容、缩容过程中,pod的生成或删除操作也是有顺序;并不会更改pod,这就是StatefulSet的特点!