Ingress`其实就是从 kuberenets 集群外部访问集群的一个入口，将外部的请求转发到集群内不同的 Service 上，其实就相当于 nginx、haproxy 等负载均衡代理服务器,Ingress controller 可以理解为一个监听器，通过不断地与 kube-apiserver 打交道，实时的感知后端 service、pod 的变化，当得到这些变化信息后，Ingress controller 再结合 Ingress 的配置，更新反向代理负载均衡器，达到服务发现的作用.

在Kubernetes 目前 提供了以下几种方案：

NodePort

LoadBalancer

Ingress

使用ingress-nginx，先创建pod服务，默认yaml文件

  1. wget  https://raw.githubusercontent.com/kubernetes/ingress-nginx/master/deploy/mandatory.yaml

vim mandatory.yaml

  1. apiVersion: v1
  2. kind: Namespace
  3. metadata:
  4.   name: ingress-nginx
  6. ---
  8. apiVersion: extensions/v1beta1
  9. kind: Deployment
  10. metadata:
  11.   name: default-http-backend
  12.   labels:
  13.     app.kubernetes.io/name: default-http-backend
  14.     app.kubernetes.io/part-of: ingress-nginx
  15.   namespace: ingress-nginx
  16. spec:
  17.   replicas: 1
  18.   selector:
  19.     matchLabels:
  20.       app.kubernetes.io/name: default-http-backend
  21.       app.kubernetes.io/part-of: ingress-nginx
  22.   template:
  23.     metadata:
  24.       labels:
  25.         app.kubernetes.io/name: default-http-backend
  26.         app.kubernetes.io/part-of: ingress-nginx
  27.     spec:
  28.       terminationGracePeriodSeconds: 60
  29.       containers:
  30.         - name: default-http-backend
  31.           # Any image is permissible as long as:
  32.           # 1. It serves a 404 page at /
  33.           # 2. It serves 200 on a /healthz endpoint
  34.           image: k8s.gcr.io/defaultbackend-amd64:1.5
  35.           livenessProbe:
  36.             httpGet:
  37.               path: /healthz
  38.               port: 8080
  39.               scheme: HTTP
  40.             initialDelaySeconds: 30
  41.             timeoutSeconds: 5
  42.           ports:
  43.             - containerPort: 8080
  44.           resources:
  45.             limits:
  46.               cpu: 10m
  47.               memory: 20Mi
  48.             requests:
  49.               cpu: 10m
  50.               memory: 20Mi
  52. ---
  53. apiVersion: v1
  54. kind: Service
  55. metadata:
  56.   name: default-http-backend
  57.   namespace: ingress-nginx
  58.   labels:
  59.     app.kubernetes.io/name: default-http-backend
  60.     app.kubernetes.io/part-of: ingress-nginx
  61. spec:
  62.   ports:
  63.     - port: 80
  64.       targetPort: 8080
  65.   selector:
  66.     app.kubernetes.io/name: default-http-backend
  67.     app.kubernetes.io/part-of: ingress-nginx
  69. ---
  71. kind: ConfigMap
  72. apiVersion: v1
  73. metadata:
  74.   name: nginx-configuration
  75.   namespace: ingress-nginx
  76.   labels:
  77.     app.kubernetes.io/name: ingress-nginx
  78.     app.kubernetes.io/part-of: ingress-nginx
  80. ---
  82. kind: ConfigMap
  83. apiVersion: v1
  84. metadata:
  85.   name: tcp-services
  86.   namespace: ingress-nginx
  87.   labels:
  88.     app.kubernetes.io/name: ingress-nginx
  89.     app.kubernetes.io/part-of: ingress-nginx
  91. ---
  93. kind: ConfigMap
  94. apiVersion: v1
  95. metadata:
  96.   name: udp-services
  97.   namespace: ingress-nginx
  98.   labels:
  99.     app.kubernetes.io/name: ingress-nginx
  100.     app.kubernetes.io/part-of: ingress-nginx
  102. ---
  104. apiVersion: v1
  105. kind: ServiceAccount
  106. metadata:
  107.   name: nginx-ingress-serviceaccount
  108.   namespace: ingress-nginx
  109.   labels:
  110.     app.kubernetes.io/name: ingress-nginx
  111.     app.kubernetes.io/part-of: ingress-nginx
  113. ---
  114. apiVersion: rbac.authorization.k8s.io/v1beta1
  115. kind: ClusterRole
  116. metadata:
  117.   name: nginx-ingress-clusterrole
  118.   labels:
  119.     app.kubernetes.io/name: ingress-nginx
  120.     app.kubernetes.io/part-of: ingress-nginx
  121. rules:
  122.   - apiGroups:
  123.       - ""
  124.     resources:
  125.       - configmaps
  126.       - endpoints
  127.       - nodes
  128.       - pods
  129.       - secrets
  130.     verbs:
  131.       - list
  132.       - watch
  133.   - apiGroups:
  134.       - ""
  135.     resources:
  136.       - nodes
  137.     verbs:
  138.       - get
  139.   - apiGroups:
  140.       - ""
  141.     resources:
  142.       - services
  143.     verbs:
  144.       - get
  145.       - list
  146.       - watch
  147.   - apiGroups:
  148.       - "extensions"
  149.     resources:
  150.       - ingresses
  151.     verbs:
  152.       - get
  153.       - list
  154.       - watch
  155.   - apiGroups:
  156.       - ""
  157.     resources:
  158.       - events
  159.     verbs:
  160.       - create
  161.       - patch
  162.   - apiGroups:
  163.       - "extensions"
  164.     resources:
  165.       - ingresses/status
  166.     verbs:
  167.       - update
  169. ---
  170. apiVersion: rbac.authorization.k8s.io/v1beta1
  171. kind: Role
  172. metadata:
  173.   name: nginx-ingress-role
  174.   namespace: ingress-nginx
  175.   labels:
  176.     app.kubernetes.io/name: ingress-nginx
  177.     app.kubernetes.io/part-of: ingress-nginx
  178. rules:
  179.   - apiGroups:
  180.       - ""
  181.     resources:
  182.       - configmaps
  183.       - pods
  184.       - secrets
  185.       - namespaces
  186.     verbs:
  187.       - get
  188.   - apiGroups:
  189.       - ""
  190.     resources:
  191.       - configmaps
  192.     resourceNames:
  193.       # Defaults to "<election-id>-<ingress-class>"
  194.       # Here: "<ingress-controller-leader>-<nginx>"
  195.       # This has to be adapted if you change either parameter
  196.       # when launching the nginx-ingress-controller.
  197.       - "ingress-controller-leader-nginx"
  198.     verbs:
  199.       - get
  200.       - update
  201.   - apiGroups:
  202.       - ""
  203.     resources:
  204.       - configmaps
  205.     verbs:
  206.       - create
  207.   - apiGroups:
  208.       - ""
  209.     resources:
  210.       - endpoints
  211.     verbs:
  212.       - get
  214. ---
  215. apiVersion: rbac.authorization.k8s.io/v1beta1
  216. kind: RoleBinding
  217. metadata:
  218.   name: nginx-ingress-role-nisa-binding
  219.   namespace: ingress-nginx
  220.   labels:
  221.     app.kubernetes.io/name: ingress-nginx
  222.     app.kubernetes.io/part-of: ingress-nginx
  223. roleRef:
  224.   apiGroup: rbac.authorization.k8s.io
  225.   kind: Role
  226.   name: nginx-ingress-role
  227. subjects:
  228.   - kind: ServiceAccount
  229.     name: nginx-ingress-serviceaccount
  230.     namespace: ingress-nginx
  232. ---
  233. apiVersion: rbac.authorization.k8s.io/v1beta1
  234. kind: ClusterRoleBinding
  235. metadata:
  236.   name: nginx-ingress-clusterrole-nisa-binding
  237.   labels:
  238.     app.kubernetes.io/name: ingress-nginx
  239.     app.kubernetes.io/part-of: ingress-nginx
  240. roleRef:
  241.   apiGroup: rbac.authorization.k8s.io
  242.   kind: ClusterRole
  243.   name: nginx-ingress-clusterrole
  244. subjects:
  245.   - kind: ServiceAccount
  246.     name: nginx-ingress-serviceaccount
  247.     namespace: ingress-nginx
  249. ---
  251. apiVersion: extensions/v1beta1
  252. kind: Deployment
  253. metadata:
  254.   name: nginx-ingress-controller
  255.   namespace: ingress-nginx
  256.   labels:
  257.     app.kubernetes.io/name: ingress-nginx
  258.     app.kubernetes.io/part-of: ingress-nginx
  259. spec:
  260.   replicas: 1
  261.   selector:
  262.     matchLabels:
  263.       app.kubernetes.io/name: ingress-nginx
  264.       app.kubernetes.io/part-of: ingress-nginx
  265.   template:
  266.     metadata:
  267.       labels:
  268.         app.kubernetes.io/name: ingress-nginx
  269.         app.kubernetes.io/part-of: ingress-nginx
  270.       annotations:
  271.         prometheus.io/port: "10254"
  272.         prometheus.io/scrape: "true"
  273.     spec:
  274.       serviceAccountName: nginx-ingress-serviceaccount
  275.       hostNetwork: true
  276.       nodeSelector:
  277.         app: ingress
  278.       containers:
  279.         - name: nginx-ingress-controller
  280.           image: quay.io/kubernetes-ingress-controller/nginx-ingress-controller:0.20.0
  281.           args:
  282.             - /nginx-ingress-controller
  283.             - --default-backend-service=$(POD_NAMESPACE)/default-http-backend
  284.             - --configmap=$(POD_NAMESPACE)/nginx-configuration
  285.             - --tcp-services-configmap=$(POD_NAMESPACE)/tcp-services
  286.             - --udp-services-configmap=$(POD_NAMESPACE)/udp-services
  287.             - --publish-service=$(POD_NAMESPACE)/ingress-nginx
  288.             - --annotations-prefix=nginx.ingress.kubernetes.io
  289.           securityContext:
  290.             capabilities:
  291.               drop:
  292.                 - ALL
  293.               add:
  294.                 - NET_BIND_SERVICE
  295.             # www-data -> 33
  296.             runAsUser: 33
  297.           env:
  298.             - name: POD_NAME
  299.               valueFrom:
  300.                 fieldRef:
  301.                   fieldPath: metadata.name
  302.             - name: POD_NAMESPACE
  303.               valueFrom:
  304.                 fieldRef:
  305.                   fieldPath: metadata.namespace
  306.           ports:
  307.             - name: http
  308.               containerPort: 80
  309.             - name: https
  310.               containerPort: 443
  311.           livenessProbe:
  312.             failureThreshold: 3
  313.             httpGet:
  314.               path: /healthz
  315.               port: 10254
  316.               scheme: HTTP
  317.             initialDelaySeconds: 10
  318.             periodSeconds: 10
  319.             successThreshold: 1
  320.             timeoutSeconds: 1
  321.           readinessProbe:
  322.             failureThreshold: 3
  323.             httpGet:
  324.               path: /healthz
  325.               port: 10254
  326.               scheme: HTTP
  327.             periodSeconds: 10
  328.             successThreshold: 1
  329.             timeoutSeconds: 1
  330. ---

将ingress固定漂移到node02节点上

  1. [root@master ingress-nginx]# kubectl get no
  2. NAME      STATUS    ROLES     AGE       VERSION
  3. master    Ready     master    9d        v1.10.0
  4. node02    Ready     <none>    8d        v1.10.0

给node02节点打上lable标签,设置hostNetwork: true提高网络访问效率

  1. kubectl label node node02 app=ingress

yaml文件配置好，lable标签也打好开始创建ingress

  1. kubectl apply -f mandatory.yaml

查看pod状态

  1. [root@master ingress-nginx]# kubectl get pod -n ingress-nginx
  2. NAME                                        READY     STATUS    RESTARTS   AGE
  3. default-http-backend-66c4fbf5b4-vkrzq       1/1       Running   4          21h
  4. nginx-ingress-controller-658cd45944-tqgvm   1/1       Running   1          21h
  1. # default
  2. 如果外界访问的域名不存在的话，则默认转发到default-http-backend这个Service，其会直接返回404：

创建i一个ingress-demo测试一下,首先创建tomcat

  1. ---
  2. # deployment
  3. apiVersion: apps/v1
  4. kind: Deployment
  5. metadata:
  6.   name: tomcat-demo
  7. spec:
  8.   selector:
  9.     matchLabels:
  10.       app: tomcat-demo
  11.   replicas: 1
  12.   template:
  13.     metadata:
  14.       labels:
  15.         app: tomcat-demo
  16.     spec:
  17.       containers:
  18.       - name: tomcat-demo
  19.         image: registry.cn-hangzhou.aliyuncs.com/liuyi01/tomcat:8.0.51-alpine
  20.         ports:
  21.         - containerPort: 8080
  23. ---
  24. # service
  25. apiVersion: v1
  26. kind: Service
  27. metadata:
  28.   name: tomcat-demo
  29. spec:
  30.   ports:
  31.   - port: 80
  32.     protocol: TCP
  33.     targetPort: 8080
  34.   selector:
  35.     app: tomcat-demo
  37. ---
  38. #ingress
  39. apiVersion: extensions/v1beta1
  40. kind: Ingress
  41. metadata:
  42.   name: tomcat-demo
  43. spec:
  44.   rules:
  45.   - host: ingress1.com
  46.     http:
  47.       paths:
  48.       - path: /
  49.         backend:
  50.           serviceName: tomcat-demo
  51.           servicePort: 8080

在创建一个nginx服务测试

  1. # deployment
  2. apiVersion: apps/v1
  3. kind: Deployment
  4. metadata:
  5.   name: tomcat-demo
  6. spec:
  7.   selector:
  8.     matchLabels:
  9.       app: tomcat-demo
  10.   replicas: 1
  11.   template:
  12.     metadata:
  13.       labels:
  14.         app: tomcat-demo
  15.     spec:
  16.       containers:
  17.       - name: tomcat-demo
  18.         image: registry.cn-hangzhou.aliyuncs.com/liuyi01/tomcat:8.0.51-alpine
  19.         ports:
  20.         - containerPort: 8080
  22. ---
  23. # service
  24. apiVersion: v1
  25. kind: Service
  26. metadata:
  27.   name: tomcat-demo
  28. spec:
  29.   ports:
  30.   - port: 80
  31.     protocol: TCP
  32.     targetPort: 8080
  33.   selector:
  34.     app: tomcat-demo
  36. ---
  37. #ingress
  38. apiVersion: extensions/v1beta1
  39. kind: Ingress
  40. metadata:
  41.   name: tomcat-demo
  42. spec:
  43.   rules:
  44.   - host: ingress2.com
  45.     http:
  46.       paths:
  47.       - path: /
  48.         backend:
  49.           serviceName: tomcat-demo
  50.           servicePort: 8080

创建nginx和tomcat pod服务

  1. [root@master nginx-ingress]# kubectl apply -f .
  2. # 查看pod情况，服务都起来了
  3. [root@master nginx-ingress]# kubectl get pod
  4. NAME                              READY     STATUS    RESTARTS   AGE
  5. load-generator-7bbbb4fdd4-l5vmz   1/1       Running   9          8d
  6. myapp-deploy-6b9c9f847c-d6fsp     1/1       Running   0          2h
  7. mynginx-85f8f5d89c-mhhzc          1/1       Running   0          2h
  8. nginx-demo-6c5d64fd5f-xnscz       1/1       Running   0          1h
  9. testservice-796fbc64bb-bczbs      1/1       Running   8          7d
  10. tomcat-demo-686b48d96c-zxfpt      1/1       Running   0          2h

配置域名解析，当前测试环境我们使用hosts文件进行解析

  1. 192.168.14.32  ingress1.com
  2. 192.168.14.32  ingress2.com

接下来通过不同域名去访问，得到不同服务

外部服务发现-ingress的更多相关文章

  1. Kubernetes之服务发现及负载Services

    Service 概述 kubernetes 中的pod是有生生灭灭的,时刻都有可能被新的pod所代替,而不可复活(pod的生命周期).一旦一个pod生命终止,通过ReplicaSets动态创建和销毁p ...

  2. Kubernetes服务发现之Service详解

    一.引子 Kubernetes Pod 是有生命周期的,它们可以被创建,也可以被销毁,然后一旦被销毁生命就永远结束.通过ReplicationController 能够动态地创建和销毁Pod(列如,需 ...

  3. (转) Docker - Docker1.12服务发现，负载均衡和Routing Mesh

    看到一篇介绍 Docker swarm以及如何编排的好文章,挪放到这里,自己学习的同时也分享出来. 原文链接: http://wwwbuild.net/dockerone/414200.html -- ...

  4. Docker与k8s的恩怨情仇（七）—— “服务发现”大法让你的内外交互原地起飞

    转载请注明出处:葡萄城官网,葡萄城为开发者提供专业的开发工具.解决方案和服务,赋能开发者. 第一章:Docker与k8s的恩怨情仇(一)-成为PaaS前浪的Cloud Foundry 第二章:Dock ...

  5. Kubernetes如何使用kube-dns实现服务发现

    大纲: •       Kubernetes中如何发现服务 •       如何发现Pod提供的服务 •       如何使用Service发现服务 •       如何使用kube-dns发现服务 ...

  6. SpringCloud接入EDAS——服务发现篇

    旁白 很久没有写技术文章了,最近不是写水文就是写小说.说到底,还是最近很少研究技术的缘故,已经到了江郎才尽的地步了. 不过,LZ无意间看到自己团队的小伙伴写的一些文章,觉得还是不错的,于是便动了心思, ...

  7. asp.net core系列 61 Ocelot 构建服务发现简单示例

    一.概述 Ocelot允许指定服务发现提供程序,如Consul或Eureka. 这二个中间件是用来实现:服务治理或秒服务发现,服务发现查找Ocelot正在转发请求的下游服务的主机和端口.目前Ocelo ...

  8. Ocelot + Consul + Registrator 基于Docker 实现服务发现、服务自动注册

    目录 1. Consul集群搭建 1.1 F&Q Consul官方推荐的host网络模式运行 2. Registrator服务注册工具 2.1 F&Q Registrator悬挂服务 ...

  9. Presto服务发现(Discovery Service)

    Presto 集群配置不管是coordinator还是worker配置项中都有一项discovery.uri,这个是一个比较核心的东西,简单来说就是服务发现的地址. coordinator和worke ...

随机推荐

  1. (原)理解码率控制模式(x264,x265,vpx)

    理解码率控制模式(x264,x265,vpx) 原文链接:https://slhck.info/video/2017/03/01/rate-control.html 翻译:lihaiping1603@ ...

  2. XT交易所API

    HTTPAPI xt为用户提供了一个简单的而又强大的API,旨在帮助用户快速高效的将xt交易功能整合到自己应用当中. API地址域名地址 域名地址:https://www.xt.com/ 使用说明 使 ...

  3. vue双循环或者多循环作用于同一元素时，在外套template标签

    如图,两个v-for,都是要循环创建tr标签.同理,v-if也是套template

  4. 浅谈 Docker 安全合规建设

    通过阅读网上帖子及浏览相关信息,大家可能会产生一种错觉:Docker 安全性不足,对 Docker 导入生产环境持保守态度.不过实际情况是,虽然我们需要对容器的安全性高度关注,但只要使用得当,完全可以 ...

  5. 浅谈Java Object

    在Java中,所有的类都继承自Object类,因此万物皆对象?也没错! 那有人会问,我的子类继承的是父类不是Object,怎么说? 如果一个类没用显示的继承某一个类,那么他就会隐式的继承 Object ...

  6. 关于vuecli的一些问题

    在vue打包之后,我们引入的css路径和js路径会变成绝对路径 需要在vue.config.js里面设置publicpath为"./" 同时在做前后端分离开发时,我们通常会用到ax ...

  7. vue2.0版本中v-html中过滤器的使用

    Vue 2.0 不再支持在 v-html 中使用过滤器 解决方法: 1:全局方法(推荐) 2:computed 属性 3:$options.filters(推荐) 1:使用全局方法: 可以在 Vue ...

  8. 论DOM中文档和元素的位置大小属性及其区别

    element.offsetLeft/Top  获取元素相对于最近的有定位的父元素的坐标,如果没有有定位的父元素,则是文档坐标 element.scrollTop/Left 获取元素滚动卷去的距离 e ...

  9. timeout超时时长优化和hystrix dashboard可视化分布式系统

    在生产环境中部署一个短路器,一开始需要将一些关键配置设置的大一些,比如timeout超时时长,线程池大小,或信号量容量 然后逐渐优化这些配置,直到在一个生产系统中运作良好 (1)一开始先不要设置tim ...

  10. 关键字ref、out

    通常,变量作为参数进行传递时,不论在方法内进行了什么操作,其原始初始化的值都不会被影响: 例如: public void TestFun1() { ; TestFun2(arg); Console.W ...