一、API 资源对象Deployment
Deployment YANL示例
vim nginx-deploy.yaml
apiVersion: apps/v1
kind: Deployment
metadata:
labels:
app: myng
name: ng-deploy
spec:
replicas: 2 ##副本数
selector:
matchLabels:
app: myng
template:
metadata:
labels:
app: myng
spec:
containers:
- name: myng
image: nginx:1.25.2
ports:
- name: myng-port
containerPort: 80使用YAML创建deploy：
# kubectl apply -f nginx-deploy.yaml
deployment.apps/ng-deploy created查看：
# kubectl get deploy |grep ng
ng-deploy 2/2 2 2 72s
nginxdp 1/1 1 1 36h
# kubectl get pod |grep deploy
ng-deploy-7b7ff4f9bc-d2g65 1/1 Running 0 103s
ng-deploy-7b7ff4f9bc-s27c9 1/1 Running 0 103s查看pod分配到哪个节点
# kubectl get po -o wide |grep deploy
ng-deploy-7b7ff4f9bc-d2g65 1/1 Running 0 2m59s 10.244.154.21 node-1-233 <none> <none>
ng-deploy-7b7ff4f9bc-s27c9 1/1 Running 0 2m59s 10.244.167.155 node-1-231 <none> <none>
二、API资源对象Service
Service检查（svc）YAML示例：
vim ng-svc.yaml
apiVersion: v1
kind: Service
metadata:
name: ngx-svc
spec:
selector:
app: myng
ports:
- protocol: TCP
port: 8080 ##service的port
targetPort: 80 ##pod的port使用YAML创建service：
# kubectl apply -f ng-svc.yaml
service/ngx-svc created查看：
# kubectl get svc |grep ngx
ngx-svc ClusterIP 10.97.206.107 <none> 8080/TCP 20s
三种Service类型
1）ClusterIP
该方式为默认类型，默认不用定义type字段。
spec:
selector:
app: myng
type: ClusterIP
ports:
- protocol: TCP
port: 8080 #service的port
targetPort: 80 #pod的port2）NodePort
如果想直接通过k8s节点的IP访问service对应的资源，可使用NodePort，NodePort对应的端口范围：30000-32767
spec:
selector:
app: myng
type: NodePort
ports:
- protocol: TCP
port: 8080 #service的port
targetPort: 80 #pod的port
nodePort: 30009 #可以自定义，也可以不定义，它会自动获取一个端口3）LoadBlancer
该方式需要配合公有云资源实现（比如：阿里云、腾讯云），需要一个公网IP作为入口，然后来负载均衡所有Pod
spec:
selector:
app: myng
type: LoadBlancer
ports:
- protocol: TCP
port: 8080 #service的port
targetPort: 80 #pod的port三、API资源对象DaemonSet
有些场景需要在每个Node上运行Pod（比如flannel插件、calico日志搜集程序）,Deployment无法做到，而Daemonet（简称ds）可以。Daemonset目标是在集群的每一个节点只运行Pod。
Daemonset不支持使用kubectl create获取YAML模板，其实Daemonset和Deployment的差异很小，除了kind不一样，还需要去掉replica配置
vi ds-demo.yaml
apiVersion: apps/v1
kind: DaemonSet
metadata:
labels:
app: ds-demo
name: ds-demo
spec:
selector:
matchLabels:
app: ds-demo
template:
metadata:
labels:
app: ds-demo
spec:
containers:
- name: ds-demo
image: nginx:1.23.2
ports:
- name: mysql-port
containerPort: 80使用YAML创建ds
# kubectl apply -f ds-demon.yaml
daemonset.apps/ds-demo created查看
# kubectl get ds
NAME DESIRED CURRENT READY UP-TO-DATE AVAILABLE NODE SELECTOR AGE
ds-demo 3 3 3 3 3 <none> 77s
# kubectl get po
NAME READY STATUS RESTARTS AGE
ds-demo-6vt22 1/1 Running 0 80s
ds-demo-tl86h 1/1 Running 0 80s
ds-demo-vmsr9 1/1 Running 0 80s
ng-deploy-7b7ff4f9bc-d2g65 1/1 Running 0 22m
ng-deploy-7b7ff4f9bc-s27c9 1/1 Running 0 22m
nginxdp-7cf46d7445-8x6gs 1/1 Running 1 (12h ago) 36h
testpod 1/1 Running 1 (12h ago) 25h但是没有在master上启动，默认master 有限制
# kubectl get node
NAME STATUS ROLES AGE VERSION
master-1-230 Ready control-plane 18d v1.27.6
node-1-231 Ready <none> 18d v1.27.6
node-1-232 Ready <none> 18d v1.27.6
node-1-233 Ready <none> 11d v1.27.6
# kubectl describe node master-1-230 |grep -i "taint"
Taints: node-role.kubernetes.io/control-plane:NoSchedule说明：Taints叫污点，如果一个节点上有污点，则不会被调度运行pod。
但是这取决于Pod的一个属性：yoleration（容忍），即这个Pod是否能容忍目标节点是否有污点。
为解决此问题，可以在Pod上增加toleration属性。修改YAML配置如下：
# cat -n ds-demon.yaml
1 apiVersion: apps/v1
2 kind: DaemonSet
3 metadata:
4 labels:
5 app: ds-demo
6 name: ds-demo
7 spec:
8 selector:
9 matchLabels:
10 app: ds-demo
11 template:
12 metadata:
13 labels:
14 app: ds-demo
15 spec:
16 tolerations:
17 - key: node-role.kubernetes.io/control-plane
18 effect: NoSchedule
19 containers:
20 - name: ds-demo
21 image: nginx:1.23.2
22 ports:
23 - name: mysql-port
24 containerPort: 80再次运行YAML
# kubectl apply -f ds-demon.yaml
daemonset.apps/ds-demo configured验证
# kubectl get ds
NAME DESIRED CURRENT READY UP-TO-DATE AVAILABLE NODE SELECTOR AGE
ds-demo 4 4 3 1 3 <none> 9m13s
# kubectl get po -o wide
NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES
ds-demo-57qcd 1/1 Running 0 86s 10.244.167.157 node-1-231 <none> <none>
ds-demo-jzlpn 1/1 Running 0 87s 10.244.154.23 node-1-233 <none> <none>
ds-demo-m8cmz 1/1 Running 0 84s 10.244.29.38 node-1-232 <none> <none>
ds-demo-sfqnt 1/1 Running 0 112s 10.244.183.193 master-1-230 <none> <none>
ng-deploy-7b7ff4f9bc-d2g65 1/1 Running 0 31m 10.244.154.21 node-1-233 <none> <none>
ng-deploy-7b7ff4f9bc-s27c9 1/1 Running 0 31m 10.244.167.155 node-1-231 <none> <none>
nginxdp-7cf46d7445-8x6gs 1/1 Running 1 (12h ago) 36h 10.244.154.20 node-1-233 <none> <none>
testpod 1/1 Running 1 (12h ago) 25h 10.244.167.153 node-1-231 <none> <none>四、API资源对象StatefulSet
Pod根据是否有数据存储分为有状态和无状态：
- 无状态：指Pod运行期间会不会产生重要数据，即使有数据产生，这些数据丢失也不影响整个应用。比如Nginx，Tomcat等应用适合无状态。
- 有状态：指Pod运行期间会产生重要数据，这些数据必须持久化，比如MySql、RedisRabbitMQ
Deployment和Daemonset适合做无状态，而有状态就是Statefulset（简称sts）
说明StatefulSet涉及数据持久化，用的StorageClass。需要先创建一个基于NFS的StorageClass
4.1 在master节点搭建NFS
NFS服务器IP地址为192.168.1.230，共享目录为/data/nfs
安装nfs
yum install -y nfs-utils nfs-server rpcbind创建共享目录，启动nfs服务
#mkdir -p /data/nfs
#echo "/data/nfs/ *(insecure,rw,sync,no_root_squash)" >> /etc/exports
# echo "/data/nfs/ *(insecure,rw,sync,no_root_squash)" >> /etc/exports
# cat /etc/exports
/data/nfs/ *(insecure,rw,sync,no_root_squash)
#systemctl enable rpcbind
#systemctl enable nfs-server
#systemctl start rpcbind
#systemctl start nfs-servernfs 配置生效命令：exportfs -r
4.2 安装SC
使用NFS的sc需要安装NFS provisioner，可以自动创建NFS的pv。使用gitee 导入 https://github.com/kubernetes-sigs/nfs-subdir-external-provisioner
下载源代码
# git clone https://gitee.com/ikubernetesi/nfs-subdir-external-provisioner.git
正克隆到 'nfs-subdir-external-provisioner'...
remote: Enumerating objects: 7784, done.
remote: Counting objects: 100% (7784/7784), done.
remote: Compressing objects: 100% (3052/3052), done.
remote: Total 7784 (delta 4443), reused 7784 (delta 4443), pack-reused 0
接收对象中: 100% (7784/7784), 8.44 MiB | 1.57 MiB/s, done.
处理 delta 中: 100% (4443/4443), done.# cd nfs-subdir-external-provisioner/deploy/
# sed -i 's/namespace: default/namespace: kube-system/' rbac.yaml
# kubectl apply -f rbac.yaml
serviceaccount/nfs-client-provisioner created
clusterrole.rbac.authorization.k8s.io/nfs-client-provisioner-runner created
clusterrolebinding.rbac.authorization.k8s.io/run-nfs-client-provisioner created
role.rbac.authorization.k8s.io/leader-locking-nfs-client-provisioner created
rolebinding.rbac.authorization.k8s.io/leader-locking-nfs-client-provisioner created修改deployment.yaml
sed -i 's/namespace: default/namespace: kube-system/' deployment.yaml ##修改命名空间为kube-system修改nfs-subdir-external-provisioner镜像仓库和nfs地址
cat deployment.yaml
# cat deployment.yaml
apiVersion: apps/v1
kind: Deployment
metadata:
name: nfs-client-provisioner
labels:
app: nfs-client-provisioner
# replace with namespace where provisioner is deployed
namespace: default
spec:
replicas: 1
strategy:
type: Recreate
selector:
matchLabels:
app: nfs-client-provisioner
template:
metadata:
labels:
app: nfs-client-provisioner
spec:
serviceAccountName: nfs-client-provisioner
containers:
- name: nfs-client-provisioner
image: registry.cn-hangzhou.aliyuncs.com/kavin1028/nfs-subdir-external-provisioner:v4.0.2
#image: registry.k8s.io/sig-storage/nfs-subdir-external-provisioner:v4.0.2
volumeMounts:
- name: nfs-client-root
mountPath: /persistentvolumes
env:
- name: PROVISIONER_NAME
value: k8s-sigs.io/nfs-subdir-external-provisioner
- name: NFS_SERVER
value: 192.168.1.230
- name: NFS_PATH
value: /data/nfs
volumes:
- name: nfs-client-root
nfs:
server: 192.168.1.230
path: /data/nfs应用YAML
# kubectl apply -f deployment.yaml
deployment.apps/nfs-client-provisioner created
# kubectl apply -f class.yaml
storageclass.storage.k8s.io/nfs-client createdSC YAML示例
# cat class.yaml
apiVersion: storage.k8s.io/v1
kind: StorageClass
metadata:
name: nfs-client
provisioner: k8s-sigs.io/nfs-subdir-external-provisioner # or choose another name, must match deployment's env PROVISIONER_NAME'
parameters:
archiveOnDelete: "false"sts 示例
vim redis-sts.yaml
apiVersion: apps/v1
kind: StatefulSet
metadata:
name: redis-sts
spec:
serviceName: redis-svc ##这里要有一个serviceName，Sts必须和service关联
volumeClaimTemplates:
- metadata:
name: redis-pvc
spec:
storageClassName: nfs-client
accessModes:
- ReadWriteMany
resources:
requests:
storage: 500Mi
replicas: 2
selector:
matchLabels:
app: redis-sts
template:
metadata:
labels:
app: redis-sts
spec:
containers:
- image: redis:6.2
name: redis
ports:
- containerPort: 6379
volumeMounts:
- name: redis-pvc
mountPath: /datavi redis-svc.yaml
apiVersion: v1
kind: Service
metadata:
name: redis-svc
spec:
selector:
app: redis-sts
ports:
- port: 6379
protocol: TCP
targetPort: 6379应用YAML 文件
# kubectl apply -f redis-sts.yaml -f redis-svc.yaml
statefulset.apps/redis-sts created
service/redis-svc created对于Sts的Pod，有如下特点：
- Pod名固定有序，后缀从0开始
- 域名固定，格式为：Pod名.Svc名，例如 redis-sts-0.redis-svc
- 每个Pod对应的PVC也是固定
验证：
ping 域名
#kubectl exec -it redis-sts-0 -- bash ##进去可以ping redis-sts-0.redis-svc 和 redis-sts-1.redis-svc
## apt-get update
# apt-get install iputils-ping -y 
创建key
kubectl exec -it redis-sts-0 -- redis-cli
127.0.0.1:6379> set name 'zhang san'
OK
127.0.0.1:6379> set age '18'
OK
模拟故障
删除后，它会自动重新创建同名Pod，再次进入查看redis key
kubectl exec -it redis-sts-0 -- redis-cli
127.0.0.1:6379> get name
"zhangsan"
127.0.0.1:6379> get age
"18"数据没有丢失
