Kubernetes之 Service & Ingress

Service

在K8S中,Pod是应用程序的载体,我们可以通过Pod来访问内部的应用程序,但是Pod又会随着伸缩容,IP又不是固定的,这就意味这不能用IP的方式进行访问

为了解决这个问题,K8S便提供Service这个资源,Service代理了同一个服务的多个Pod集合,对外表现为一个访问入口,访问该入口的请求将经过负载均衡,转发到后端 Pod 中的容器。

kube-proxy

Service在很多情况下只是一个概念,真正起作用的其实是kube-proxy服务进程,每个Node节点上都运行着一个kube-proxy服务进程。当创建Service的时候会通过api-server向etcd写入创建的service的信息,而kube-proxy会基于监听的机制发现这种Service的变动,然后它会将最新的Service信息转换成对应的访问规则。

目前Service支持三种工作模式,我们可以修改

# 开启ipvs 编辑资源清单文件,修改mode: "ipvs"
[root@master /]# kubectl edit cm kube-proxy -n kube-system
configmap/kube-proxy edited
​
# 获取现在已有的服务
[root@master /]# kubectl get pod -l k8s-app=kube-proxy -n kube-system
NAME               READY   STATUS    RESTARTS   AGE
kube-proxy-4pc5n   1/1     Running   0          2d2h
kube-proxy-gw67r   1/1     Running   0          2d3h
kube-proxy-rqbgm   1/1     Running   0          2d2h
​
# 删除现在已有的服务
[root@master /]# kubectl delete pod -l k8s-app=kube-proxy -n kube-system
pod "kube-proxy-4pc5n" deleted
pod "kube-proxy-gw67r" deleted
pod "kube-proxy-rqbgm" deleted
​
# 获取删除后自动重启的服务
[root@master /]# kubectl get pod -l k8s-app=kube-proxy -n kube-system
NAME               READY   STATUS    RESTARTS   AGE
kube-proxy-4rl6z   1/1     Running   0          5s
kube-proxy-65bt7   1/1     Running   0          5s
kube-proxy-w94rl   1/1     Running   0          4s

userspace 模式

该模式下kube-proxy会创建一个监听端口 ,发向Cluster IP的请求被Iptables规则重定向到kube-proxy监听的端口上

kube-proxy根据LB算法选择一个提供服务的Pod并和其建立链接,以将请求转发到Pod上

该模式下,kube-proxy充当了一个四层负责均衡器的角色。由于kube-proxy运行在userspace中,在进行转发处理时会增加内核和用户空间之间的数据拷贝,虽然比较稳定,但是效率比较低

iptables 模式

该模式下,kube-proxy为service映射的每个Pod创建对应的iptables规则,直接将发向Cluster IP的请求重定向到一个Pod IP。 该模式下kube-proxy不承担四层负责均衡器的角色,只负责创建iptables规则。该模式的优点是较userspace模式效率更高,但不能提供灵活的LB策略,当后端Pod不可用时也无法进行重试

ipvs 模式

ipvs模式和iptables类似,kube-proxy监控Pod的变化并创建相应的ipvs规则。ipvs相对iptables转发效率更高。除此以外,ipvs支持更多的LB算法。

Pod和Service的关系

之前我们已经说到Controller和Pod之间是如何建立关系的。Service与Pod取得联系的方式如出一辙

img

常用的Service类型

  • kubectl expose --help 查看服务的暴露帮助文档,在最下可以看到几个关键字眼

img

  1. ClusterIP

    • 一般用于集群的内部使用,对外无效,默认的也就是这种类型

  2. NodePort

    • 对外访问应用时使用,比如我们的前端Pod是要对外暴露的,面向集群外部的访问

  3. LoadBalancer

    • 使用外接负载均衡器完成到服务的负载分发,注意此模式需要外部云环境支持

  4. ExternalName

    • 把集群外部的服务引入集群内部,直接使用

测试环境,在演示之前先把下面的测试环境搭建起来

资源清单文件:nginx-deploy.yaml

apiVersion: apps/v1
kind: Deployment      
metadata:
  name: nginx-deploy
  name
spec: 
  replicas: 3
  selector:
    matchLabels:
      app: nginx-pod
  template:
    metadata:
      labels:
        app: nginx-pod
    spec:
      containers:
      - name: nginx
        image: nginx:1.17.1
        ports:
        - containerPort: 80
[root@master k8s]# kubectl apply -f nginx-deploy.yaml
deployment.apps/nginx-deploy created
# 可以看到除了自带的哪个ClusterIP , 没有其他的Service被创建
[root@master k8s]# kubectl get svc
NAME         TYPE        CLUSTER-IP   EXTERNAL-IP   PORT(S)   AGE
kubernetes   ClusterIP   10.96.0.1    <none>        443/TCP   2d4h
​
#三个副本正在缓缓启动 注意看 IP 项
[root@master k8s]# kubectl get pods -o wide
NAME                            READY   STATUS    RESTARTS   AGE   IP           NODE    NOMINATED NODE   READINESS GATES
nginx-deploy-7d7dd5499b-7nfkz   1/1     Running   0          19m   10.244.1.5   node1   <none>           <none>
nginx-deploy-7d7dd5499b-9c5lf   1/1     Running   0          19m   10.244.1.6   node1   <none>           <none>
nginx-deploy-7d7dd5499b-lwcq6   1/1     Running   0          19m   10.244.2.6   node2   <none>           <none>
​
# 依次进入三个容器
[root@master k8s]# kubectl exec -it nginx-deploy-7d7dd5499b-7nfkz -n dev /bin/sh
​
#一台分配一句填充一下nginx的index.html
echo "master hello!" > /usr/share/nginx/html/index.html
echo "master master hello!" > /usr/share/nginx/html/index.html
echo "master  master masterhello!" > /usr/share/nginx/html/index.html
​
# 可以看到 , 我们通过 kubectl get pods -o wide 获取到的IP依次访问了这些Nginx
[root@master k8s]# curl 10.244.1.5
master hello !
[root@master k8s]# curl 10.244.1.6
master master hello !
[root@master k8s]# curl 10.244.2.6
master master master hello !

ClusterIP

创建一个Service的资源:service_cluster_ip.yaml

  • 这里注意:这里我们通过 selector 匹配那些 标签key = app value=nginx-pod 的资源,也就是上面我们部署的三台Nginx

apiVersion: v1
kind: Service
metadata:
  name: service-clusterip
spec:
  selector:
    app: nginx-pod
  type: ClusterIP
  ports:
  - port: 80         # Service端口       
    targetPort: 80   # 代理pod的端口
# 部署Servicer类型的资源
[root@master k8s]# kubectl apply -f service_cluster_ip.yaml
service/service-clusterip created
​
# 查看service , 可以看到 service-clusterip 这个service的 clusterIp为  10.109.230.1
[root@master k8s]# kubectl get service
NAME                TYPE        CLUSTER-IP     EXTERNAL-IP   PORT(S)   AGE
kubernetes          ClusterIP   10.96.0.1      <none>        443/TCP   5d19h
service-clusterip   ClusterIP   10.109.230.1   <none>        80/TCP    23s
​
# 查看该service的详细信息
[root@master k8s]# kubectl describe service service-clusterip
Name:              service-clusterip
Namespace:         default
Labels:            <none>
Annotations:       Selector:  app=nginx-pod
Type:              ClusterIP
IP:                10.109.230.1
Port:              <unset>  80/TCP
TargetPort:        80/TCP
Endpoints:         10.244.1.5:80,10.244.1.6:80,10.244.2.6:80
Session Affinity:  None
Events:            <none>
​
# 开始使用service暴露的clusterIP进行访问 , 可以发现经过负载均衡分发,我们访问到了三台Nginx , 但是这个地址只能再集群内部的节点上使用
[root@master k8s]# curl 10.109.230.1
master master master hello !
[root@master k8s]# curl 10.109.230.1
master master hello !
[root@master k8s]# curl 10.109.230.1
master hello !
Endedpoint
  • 大家看看service详细描述信息里有一个数据项目:Endpoints: 10.244.1.5:80,10.244.1.6:80,10.244.2.6:80

  • Endpoint是kubernetes中的一个资源对象,存储在etcd中,用来记录一个service对应的所有pod的访问地址,它是根据service配置文件中selector描述产生的。

# 获取详细描述
[root@master k8s]# kubectl describe service service-clusterip
Name:              service-clusterip
Namespace:         default
Labels:            <none>
Annotations:       Selector:  app=nginx-pod
Type:              ClusterIP
IP:                10.109.230.1
Port:              <unset>  80/TCP
TargetPort:        80/TCP
Endpoints:         10.244.1.5:80,10.244.1.6:80,10.244.2.6:80
Session Affinity:  None
Events:            <none>
​
# 查看pod的信息
[root@master k8s]# kubectl get pods -o wide
NAME                            READY   STATUS    RESTARTS   AGE     IP           NODE    NOMINATED NODE   READINESS GATES
nginx-deploy-7d7dd5499b-7nfkz   1/1     Running   0          3d15h   10.244.1.5   node1   <none>           <none>
nginx-deploy-7d7dd5499b-9c5lf   1/1     Running   0          3d15h   10.244.1.6   node1   <none>           <none>
nginx-deploy-7d7dd5499b-lwcq6   1/1     Running   0          3d15h   10.244.2.6   node2   <none>           <none>

nodePort

创建一个Service的资源:service_node_port.yaml

  • 这里注意:这里我们通过 selector 匹配那些 标签key = app value=nginx-pod 的资源,也就是上面我们部署的三台Nginx

apiVersion: v1
kind: Service
metadata:
  name: service-nodeport
spec:
  selector:
    app: nginx-pod
  type: NodePort     # service类型
  ports:
  - port: 80
    nodePort: 30003  # 指定绑定的node的端口(默认的取值范围是:30000-32767), 如果不指定,会默认分配
    targetPort: 80
[root@master k8s]# kubectl apply -f service_node_port.yaml
service/service-nodeport created
​
# 获取到刚刚部署的service-nodeport对外暴露的端口是:30003
[root@master k8s]# kubectl get service
NAME                TYPE        CLUSTER-IP     EXTERNAL-IP   PORT(S)        AGE
kubernetes          ClusterIP   10.96.0.1      <none>        443/TCP        5d20h
service-clusterip   ClusterIP   10.109.230.1   <none>        80/TCP         80m
service-nodeport    NodePort    10.103.77.9    <none>        80:30003/TCP   23s
​
# 获取详细描述
[root@master k8s]# kubectl describe service service-nodeport
Name:                     service-nodeport
Namespace:                default
Labels:                   <none>
Annotations:              Selector:  app=nginx-pod
Type:                     NodePort
IP:                       10.103.77.9
Port:                     <unset>  80/TCP
TargetPort:               80/TCP
NodePort:                 <unset>  30003/TCP
Endpoints:                10.244.1.5:80,10.244.1.6:80,10.244.2.6:80
Session Affinity:         None
External Traffic Policy:  Cluster
Events:                   <none>

此时我们再配合节点机器的Ip地址 + nodePort暴露出来的端口,通过浏览器访问即可

loadBalance

node节点一般是匿名部署的,ip地址一般不对外做暴露,若想访问,以下有两种实现方式

  1. 在集群节点中,找到一台可以外网访问的机器,安装Nginx,做反向代理,手动把可以访问的节点添加到Nginx里面

  2. LoadBalancer的介入,在基于ClusterIP和已经以NodePort开放的一个服务Service的基础上,向云提供者申请一个负载均衡器,将流量转发到已经以NodePort形式开发的服务Service上

  3. 较为麻烦,一般不做使用

ExternalName

ExternalName类型的Service用于引入集群外部的服务,它通过externalName属性指定外部一个服务的地址,然后在集群内部访问此service就可以访问到外部的服务了。

apiVersion: v1
kind: Service
metadata:
  name: service-externalname
spec:
  type: ExternalName           # service类型
  externalName: www.baidu.com  #改成ip地址也可以

Ingress

在上面我们说到集群之内服务想要暴露,只有通过nodePort或者loadBalance这两种方式,他们都有自己的缺点

  • nodePort :一个Service就要占用集群机器的一个端口,服务变多时不好管控

  • loadBalance : 每个service都需要一个lb , 麻烦之极 , 并且需要kubernetes之外设备的支持

基于这种环境下,Ingress就出现了:Ingress只需要一个NodePort或者一个LB就可以满足暴露多个Service的需求

Ingress的工作原理类似与Nginx,可以理解可以理解成在Ingress里建立诸多映射规则,Ingress Controller通过监听这些配置规则并转化成Nginx的反向代理配置 , 然后对外部提供服务

  • ingress :kubernetes中的一个对象,作用是定义请求如何转发到service的规则

  • ingress controller:具体实现反向代理及负载均衡的程序,对ingress定义的规则进行解析,实现请求转发,实现方式有很多,比如Nginx, Contour, Haproxy等

ingress-nginx

你必须拥有一个 Ingress控制器 才能满足 Ingress 的要求。 Ingress控制器-官方文档 ,ingress-nginx就是Ingress的实现方案之一,下面我们将其部署起来用作演示

创建一个命名空间,待会Ingress就部署在该命名空间下

[root@master /]# kubectl create namespace ingress-nginx

还是之前Nginx的Deployment文件

apiVersion: apps/v1
kind: Deployment      
metadata:
  name: nginx-deploy
spec: 
  replicas: 3
  selector:
    matchLabels:
      app: nginx-pod
  template:
    metadata:
      labels:
        app: nginx-pod
    spec:
      containers:
      - name: nginx
        image: nginx:1.17.1
        ports:
        - containerPort: 80
#将上面的资源部署到 ingress-nginx 这个命名空间下
[root@master /]# kubectl apply -f nginx-deploy.yaml -n ingress-nginx

还是之前Nginx的Deployment文件

apiVersion: v1
kind: Service
metadata:
  name: service-nodeport
spec:
  selector:
    app: nginx-pod
  type: NodePort     
  ports:
  - port: 80
    nodePort: 30003  #指定绑定的node的端口(默认的取值范围是:30000-32767), 如果不指定,会默认分配
    targetPort: 80
#将上面的资源部署到 ingress-nginx 这个命名空间下
[root@master /]# kubectl apply -f nginx-service.yaml -n ingress-nginx

然后开始部署 Ingress Controller

  • 这里我们使用官方维护的nginx控制器进行部署,下面你需要变动的数据可能就是命名空间了,如果你的命名空间不是 ingress-nginx , 则需要改动一二

  • 下面这个资源清单文件围绕:nginx-ingress-controller 这个Deployment 做了配套账户、权限、配置等,我们只需关注该Deployment 即可

apiVersion: v1
kind: Namespace
metadata:
  name: ingress-nginx
  labels:
    app.kubernetes.io/name: ingress-nginx
    app.kubernetes.io/part-of: ingress-nginx
​
---
​
kind: ConfigMap
apiVersion: v1
metadata:
  name: nginx-configuration
  namespace: ingress-nginx
  labels:
    app.kubernetes.io/name: ingress-nginx
    app.kubernetes.io/part-of: ingress-nginx
​
---
kind: ConfigMap
apiVersion: v1
metadata:
  name: tcp-services
  namespace: ingress-nginx
  labels:
    app.kubernetes.io/name: ingress-nginx
    app.kubernetes.io/part-of: ingress-nginx
​
---
kind: ConfigMap
apiVersion: v1
metadata:
  name: udp-services
  namespace: ingress-nginx
  labels:
    app.kubernetes.io/name: ingress-nginx
    app.kubernetes.io/part-of: ingress-nginx
​
---
apiVersion: v1
kind: ServiceAccount
metadata:
  name: nginx-ingress-serviceaccount
  namespace: ingress-nginx
  labels:
    app.kubernetes.io/name: ingress-nginx
    app.kubernetes.io/part-of: ingress-nginx
​
---
apiVersion: rbac.authorization.k8s.io/v1beta1
kind: ClusterRole
metadata:
  name: nginx-ingress-clusterrole
  labels:
    app.kubernetes.io/name: ingress-nginx
    app.kubernetes.io/part-of: ingress-nginx
rules:
  - apiGroups:
      - ""
    resources:
      - configmaps
      - endpoints
      - nodes
      - pods
      - secrets
    verbs:
      - list
      - watch
  - apiGroups:
      - ""
    resources:
      - nodes
    verbs:
      - get
  - apiGroups:
      - ""
    resources:
      - services
    verbs:
      - get
      - list
      - watch
  - apiGroups:
      - ""
    resources:
      - events
    verbs:
      - create
      - patch
  - apiGroups:
      - "extensions"
      - "networking.k8s.io"
    resources:
      - ingresses
    verbs:
      - get
      - list
      - watch
  - apiGroups:
      - "extensions"
      - "networking.k8s.io"
    resources:
      - ingresses/status
    verbs:
      - update
​
---
apiVersion: rbac.authorization.k8s.io/v1beta1
kind: Role
metadata:
  name: nginx-ingress-role
  namespace: ingress-nginx
  labels:
    app.kubernetes.io/name: ingress-nginx
    app.kubernetes.io/part-of: ingress-nginx
rules:
  - apiGroups:
      - ""
    resources:
      - configmaps
      - pods
      - secrets
      - namespaces
    verbs:
      - get
  - apiGroups:
      - ""
    resources:
      - configmaps
    resourceNames:
      # Defaults to "<election-id>-<ingress-class>"
      # Here: "<ingress-controller-leader>-<nginx>"
      # This has to be adapted if you change either parameter
      # when launching the nginx-ingress-controller.
      - "ingress-controller-leader-nginx"
    verbs:
      - get
      - update
  - apiGroups:
      - ""
    resources:
      - configmaps
    verbs:
      - create
  - apiGroups:
      - ""
    resources:
      - endpoints
    verbs:
      - get
​
---
apiVersion: rbac.authorization.k8s.io/v1beta1
kind: RoleBinding
metadata:
  name: nginx-ingress-role-nisa-binding
  namespace: ingress-nginx
  labels:
    app.kubernetes.io/name: ingress-nginx
    app.kubernetes.io/part-of: ingress-nginx
roleRef:
  apiGroup: rbac.authorization.k8s.io
  kind: Role
  name: nginx-ingress-role
subjects:
  - kind: ServiceAccount
    name: nginx-ingress-serviceaccount
    namespace: ingress-nginx
​
---
apiVersion: rbac.authorization.k8s.io/v1beta1
kind: ClusterRoleBinding
metadata:
  name: nginx-ingress-clusterrole-nisa-binding
  labels:
    app.kubernetes.io/name: ingress-nginx
    app.kubernetes.io/part-of: ingress-nginx
roleRef:
  apiGroup: rbac.authorization.k8s.io
  kind: ClusterRole
  name: nginx-ingress-clusterrole
subjects:
  - kind: ServiceAccount
    name: nginx-ingress-serviceaccount
    namespace: ingress-nginx
​
---
​
apiVersion: apps/v1
kind: Deployment
metadata:
  name: nginx-ingress-controller
  namespace: ingress-nginx
  labels:
    app.kubernetes.io/name: ingress-nginx
    app.kubernetes.io/part-of: ingress-nginx
spec:
  replicas: 1
  selector:
    matchLabels:
      app.kubernetes.io/name: ingress-nginx
      app.kubernetes.io/part-of: ingress-nginx
  template:
    metadata:
      labels:
        app.kubernetes.io/name: ingress-nginx
        app.kubernetes.io/part-of: ingress-nginx
      annotations:
        prometheus.io/port: "10254"
        prometheus.io/scrape: "true"
    spec:
      hostNetwork: true
      # wait up to five minutes for the drain of connections
      terminationGracePeriodSeconds: 300
      serviceAccountName: nginx-ingress-serviceaccount
      nodeSelector:
        kubernetes.io/os: linux
      containers:
        - name: nginx-ingress-controller
          image: lizhenliang/nginx-ingress-controller:0.30.0
          args:
            - /nginx-ingress-controller
            - --configmap=$(POD_NAMESPACE)/nginx-configuration
            - --tcp-services-configmap=$(POD_NAMESPACE)/tcp-services
            - --udp-services-configmap=$(POD_NAMESPACE)/udp-services
            - --publish-service=$(POD_NAMESPACE)/ingress-nginx
            - --annotations-prefix=nginx.ingress.kubernetes.io
          securityContext:
            allowPrivilegeEscalation: true
            capabilities:
              drop:
                - ALL
              add:
                - NET_BIND_SERVICE
            # www-data -> 101
            runAsUser: 101
          env:
            - name: POD_NAME
              valueFrom:
                fieldRef:
                  fieldPath: metadata.name
            - name: POD_NAMESPACE
              valueFrom:
                fieldRef:
                  fieldPath: metadata.namespace
          ports:
            - name: http
              containerPort: 80
              protocol: TCP
            - name: https
              containerPort: 443
              protocol: TCP
          livenessProbe:
            failureThreshold: 3
            httpGet:
              path: /healthz
              port: 10254
              scheme: HTTP
            initialDelaySeconds: 10
            periodSeconds: 10
            successThreshold: 1
            timeoutSeconds: 10
          readinessProbe:
            failureThreshold: 3
            httpGet:
              path: /healthz
              port: 10254
              scheme: HTTP
            periodSeconds: 10
            successThreshold: 1
            timeoutSeconds: 10
          lifecycle:
            preStop:
              exec:
                command:
                  - /wait-shutdown
​
---
​
apiVersion: v1
kind: LimitRange
metadata:
  name: ingress-nginx
  namespace: ingress-nginx
  labels:
    app.kubernetes.io/name: ingress-nginx
    app.kubernetes.io/part-of: ingress-nginx
spec:
  limits:
  - min:
      memory: 90Mi
      cpu: 100m
    type: Container

配置规则

# 三个Nginx的pod 已经就绪 、 一个Ingress Controller 也是以Pod的形式在运行着
[root@master ninja_download]# kubectl get pods -n ingress-nginx -o wide 
NAME                                       READY   STATUS    RESTARTS   AGE    IP                NODE    NOMINATED NODE   READINESS GATES
nginx-deploy-7d7dd5499b-bfsrs              1/1     Running   0          30m    10.244.1.7        node1   <none>           <none>
nginx-deploy-7d7dd5499b-dpn99              1/1     Running   0          30m    10.244.1.8        node1   <none>           <none>
nginx-deploy-7d7dd5499b-tvh5s              1/1     Running   0          30m    10.244.2.7        node2   <none>           <none>
nginx-ingress-controller-766fb9f77-nw7gb   1/1     Running   0          41m   192.168.239.140   node2   <none>           <none>
​
# 关联了三台Nginx的Service 也处于就绪状态
[root@master /]# kubectl get service -n ingress-nginx
NAME               TYPE       CLUSTER-IP       EXTERNAL-IP   PORT(S)        AGE
service-nodeport   NodePort   10.102.164.188   <none>        80:30003/TCP   28m

现在我们Pod通过Deployment创建好了 , 对应的Service也创建好了 , Ingress Controller 也处于就绪状态

下面我们就剩下最后一步:配置 Ingress 规则,让他们产生关联了

apiVersion: networking.k8s.io/v1beta1
kind: Ingress
metadata:
  name: ninja-ingress
spec:
  rules:
  - host: ninja.ingress.com                #对外监听并维护的域名
    http:
      paths:
      - path: /
        backend:
          serviceName: service-nodeport   # 这是关联的 service 的 serviceName
          servicePort: 80                 #监听的端口
# 将该资源部署在同一命名空间下 
[root@master /]# kubectl apply -f ingress-rule.yaml -n ingress-nginx

因为我们监听的是这个域名:ninja.ingress.com 所以我们要想访问的话只有修改一下win本地的hosts文件

  • C:\Windows\System32\drivers\etc\hosts

  • 192.168.239.140 ninja.ingress.com : 前面是我们部署Ingress Controller Pod的节点的IP , 后面是Ingress监听的域名信息

下面我们来看看Ingress资源信息

[root@master ninja_download]# kubectl get ingress -n ingress-nginx
NAME            CLASS    HOSTS               ADDRESS   PORTS   AGE
ninja-ingress   <none>   ninja.ingress.com             80      50m

工作原理

  1. 用户编写Ingress路由规则,说明哪个域名对应哪个K8S中的哪个Service

  2. Ingress Controller 动态感知 Ingress路由规则的变化,生成一段对饮的Nginx反向代理配置

  3. Ingress动态将该配置写入到一个运行中的Nginx中,并使其生效

更多路由规则详见官网

 

 

 

 

posted @ 2021-01-29 20:21  鞋破露脚尖儿  阅读(503)  评论(0编辑  收藏  举报