参考文档:
- Github:
- Kubernetes ingress:
- Ingress:
- 配置示例:
- Github示例:
- Traefik示例:
Ingress是对外服务到集群内的Service之间规则的集合:允许进入集群的请求被转发至集群内的Service。
Ingress能把Service配置成外网能够访问的url,流量负载均衡,终止ssl,提供基于域名访问的虚拟主机等,用户通过访问url访问Service。
Ingress-controller负责处理所有Ingress的请求流量,它通常是一个负载均衡器。
一.环境
1. 基础环境
组件 | 版本 | Remark |
kubernetes | v1.9.2 | |
Ingress-nginx | 0.11.0 | |
default-backend | 1.4 |
2. 原理
- ingress策略本质是转发的规则;
- ingress-controller基于ingress策略将客户端的请求转发到service对应的后端endpoint,即Pod上;实现了为所有后端service提供统一入口,基于不同的http url向后转发负载分发规则,并可以灵活设置7层的负载分发策略的功能;一般由nginx实现。
二.部署ingress-nginx
1. 准备images
kubernetes部署Pod服务时,为避免部署时发生pull镜像超时的问题,建议提前将相关镜像pull到相关所有节点(实验),或搭建本地镜像系统。
- 基础环境已做了镜像加速,可参考:
- 需要从gcr.io pull的镜像,已利用Docker Hub的"Create Auto-Build GitHub"功能(Docker Hub利用GitHub上的Dockerfile文件build镜像),在个人的Docker Hub build成功,可直接pull到本地使用。
# ingress-controller默认的backend,用于在客户端访问的url地址不存在时,能够返回一个正确的404应答[root@kubenode1 ~]# docker pull netonline/defaultbackend:1.4# ingress-nginx[root@kubenode1 ~]# docker pull netonline/nginx-ingress-controller:0.11.0
2. 下载ingress-nginx相关yaml范本
# 相关的yaml文件可在1个或多个master节点下载后修改[root@kubenode1 ~]# mkdir -p /usr/local/src/yaml/ingress[root@kubenode1 ~]# cd /usr/local/src/yaml/ingress/#下载链接: # namespace[root@kubenode1 ingress]# wget # configmap,此验证未使用[root@kubenode1 ingress]# wget # tcp-service-configmap[root@kubenode1 ingress]# wget # udp-service-configmap,此验证未使用[root@kubenode1 ingress]# wget # rbac[root@kubenode1 ingress]# wget # default-backend[root@kubenode1 ingress]# wget # with-rbac[root@kubenode1 ingress]# wget # without-rbac,此验证未使用[root@kubenode1 ingress]# wget # patch,此验证未使用[root@kubenode1 ingress]# wget
3. namespace.yaml
# ingress-nginx github文档中将Namespace,ConfingMap,ServiceAccount,Deployment,default-backend,xxx-services-configmap等服务的yaml配置文件独立保存,以下章节分别针对各yaml文件修改,红色加粗字体即修改部分;# 对Pod yaml文件的编写这里不做展开,可另参考资料,如《Kubernetes权威指南》;# 修改后的ingress-nginx相关yaml文件请见:# namespace.yaml不做修改,创建1个独立的namespace[root@kubenode1 ingress]# cat namespace.yaml apiVersion: v1kind: Namespacemetadata: name: ingress-nginx
4. tcp-services-configmap.yaml
# tcp-services-configmap.yaml不做修改[root@kubenode1 ingress]# cat tcp-services-configmap.yaml kind: ConfigMapapiVersion: v1metadata: name: tcp-services namespace: ingress-nginx
5. rbac.yaml
# ingress-controller需要监听apiserver,获取ingress定义,通过rbac授权;# rbac.yaml文件不用修改[root@kubenode1 ingress]# cat rbac.yaml apiVersion: v1kind: ServiceAccountmetadata: name: nginx-ingress-serviceaccount namespace: ingress-nginx---apiVersion: rbac.authorization.k8s.io/v1beta1kind: ClusterRolemetadata: name: nginx-ingress-clusterrolerules: - apiGroups: - "" resources: - configmaps - endpoints - nodes - pods - secrets verbs: - list - watch - apiGroups: - "" resources: - nodes verbs: - get - apiGroups: - "" resources: - services verbs: - get - list - watch - apiGroups: - "extensions" resources: - ingresses verbs: - get - list - watch - apiGroups: - "" resources: - events verbs: - create - patch - apiGroups: - "extensions" resources: - ingresses/status verbs: - update---apiVersion: rbac.authorization.k8s.io/v1beta1kind: Rolemetadata: name: nginx-ingress-role namespace: ingress-nginxrules: - apiGroups: - "" resources: - configmaps - pods - secrets - namespaces verbs: - get - apiGroups: - "" resources: - configmaps resourceNames: # Defaults to "- " # Here: " - " # 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/v1beta1kind: RoleBindingmetadata: name: nginx-ingress-role-nisa-binding namespace: ingress-nginxroleRef: apiGroup: rbac.authorization.k8s.io kind: Role name: nginx-ingress-rolesubjects: - kind: ServiceAccount name: nginx-ingress-serviceaccount namespace: ingress-nginx---apiVersion: rbac.authorization.k8s.io/v1beta1kind: ClusterRoleBindingmetadata: name: nginx-ingress-clusterrole-nisa-bindingroleRef: apiGroup: rbac.authorization.k8s.io kind: ClusterRole name: nginx-ingress-clusterrolesubjects: - kind: ServiceAccount name: nginx-ingress-serviceaccountnamespace: ingress-nginx
6. default-backend.yaml
# 提供1个默认的后台404错误页面与/healthz的健康检查页;# 含1个Deployment与1个service;# 只需要修改Pod启动调用的image文件名[root@kubenode1 ingress]# vim default-backend.yamlapiVersion: extensions/v1beta1kind: Deploymentmetadata: name: default-http-backend labels: app: default-http-backend namespace: ingress-nginxspec: replicas: 1 selector: matchLabels: app: default-http-backend template: metadata: labels: app: default-http-backend spec: terminationGracePeriodSeconds: 60 containers: - name: default-http-backend # Any image is permissible as long as: # 1. It serves a 404 page at / # 2. It serves 200 on a /healthz endpoint image: netonline/defaultbackend:1.4 livenessProbe: httpGet: path: /healthz port: 8080 scheme: HTTP initialDelaySeconds: 30 timeoutSeconds: 5 ports: - containerPort: 8080 resources: limits: cpu: 10m memory: 20Mi requests: cpu: 10m memory: 20Mi---apiVersion: v1kind: Servicemetadata: name: default-http-backend namespace: ingress-nginx labels: app: default-http-backendspec: ports: - port: 80 targetPort: 8080 selector: app: default-http-backend
7. with-rbac.yaml
Ingress-Controller以Pod的形式运行,监控apiserver的/ingress接口后端的backend services,如果service发生变化,则Ingress-Controller自动更新转发规则。
基本逻辑如下:
- 监听apiserver,获取全部ingress定义;
- 基于ingress定义,生成nginx的配置文件/etc/nginx/nginx.conf;
- 执行nginx -s reload,重新加载nginx.conf配置文件的内容。
# without-rbac.yaml与with-rbac.yaml的区别是没有调用rabc.yaml中定义的ServiceAccount: nginx-ingress-serviceaccount,这里访问apiserve需要认证;# github文档给的kind: Deployment,replicas: 1,即在1个节点上启动1个ingress-nginx controller Pod,外部流量访问该节点,由该节点负载后端services;但Pod本身会因为故障而转移,节点ip会变更,DaemonSet会在多个节点(可以利用Pod的亲和性将指定Pod部署到指点节点)生成ingress-nginx controller Pod,则客户端可以访问任意节点;或者在前端部署负载,使用vip访问后端3个节点;# hostNetwork: true,暴露ingress-nginx controller的相关业务端口到主机;# 验证中暂时用不到的服务不启用,则相应的ingress-controller中的对应参数也注释不用[root@kubenode1 ingress]# vim with-rbac.yamlapiVersion: extensions/v1beta1# kind: Deployment# 变更kindkind: DaemonSetmetadata: name: nginx-ingress-controller namespace: ingress-nginx spec: # 已变更kind,注释副本数 # replicas: 1 selector: matchLabels: app: ingress-nginx template: metadata: labels: app: ingress-nginx annotations: prometheus.io/port: '10254' prometheus.io/scrape: 'true' spec: serviceAccountName: nginx-ingress-serviceaccount # 暴露主机端口 hostNetwork: true containers: - name: nginx-ingress-controller # 变更调用image名 image: netonline/nginx-ingress-controller:0.11.0 args: - /nginx-ingress-controller - --default-backend-service=$(POD_NAMESPACE)/default-http-backend # 此验证未使用ConfigMap,注释 # - --configmap=$(POD_NAMESPACE)/nginx-configuration - --tcp-services-configmap=$(POD_NAMESPACE)/tcp-services # 此验证未使用udp-service-ConfigMap,注释 # - --udp-services-configmap=$(POD_NAMESPACE)/udp-services - --annotations-prefix=nginx.ingress.kubernetes.io env: - name: POD_NAME valueFrom: fieldRef: fieldPath: metadata.name - name: POD_NAMESPACE valueFrom: fieldRef: fieldPath: metadata.namespace ports: - name: http containerPort: 80 # 暴露的主机端口 hostPort: 80 - name: https containerPort: 443 hostPort: 443 livenessProbe: failureThreshold: 3 httpGet: path: /healthz port: 10254 scheme: HTTP initialDelaySeconds: 10 periodSeconds: 10 successThreshold: 1 timeoutSeconds: 1 readinessProbe: failureThreshold: 3 httpGet: path: /healthz port: 10254 scheme: HTTP periodSeconds: 10 successThreshold: 1 timeoutSeconds: 1
8. 启动ingress-nginx
[root@kubenode1 ingress]# kubectl create -f namespace.yaml [root@kubenode1 ingress]# kubectl create -f tcp-services-configmap.yaml [root@kubenode1 ingress]# kubectl create -f rbac.yaml [root@kubenode1 ingress]# kubectl create -f default-backend.yaml [root@kubenode1 ingress]# kubectl create -f with-rbac.yaml
9. 设置iptables
# 3台master节点均设置,with-rbac.yaml/without-rbac.yaml会启用”hostNetwork”,并且开放tcp80,443,18080(/nginx-status)端口;# 因为这里docker服务已启动,采用直接在input链追加开放端口的方式;# 建议在/etc/sysconfig/iptables配置文件中将相应端口打开;# 如果采用”service iptables save”命令会将当前已有iptables规则(含docker服务相关规则)全部写入配置文件,慎用[root@kubenode1 ~]# iptables -A INPUT -m state --state NEW -m tcp -p tcp --dport 80 -j ACCEPT[root@kubenode1 ~]# iptables -A INPUT -m state --state NEW -m tcp -p tcp --dport 443 -j ACCEPT[root@kubenode1 ~]# iptables -A INPUT -m state --state NEW -m tcp -p tcp --dport 18080 -j ACCEPT
10. 验证
# 1个default-htt-backend与3个ingress-nginx Pod已运行[root@kubenode1 ~]# kubectl get pods -n ingress-nginx -o wide
# 节点本机相应端口已被使用[root@kubenode1 ~]# netstat -tunlp | grep nginx
# 访问任意节点的80端口,返回404页面,ingress-nginx controller与default-http-backend生效[root@kubenode1 ~]# curl http://172.30.200.21[root@kubenode1 ~]# curl http://172.30.200.22[root@kubenode1 ~]# curl http://172.30.200.23
三.部署ingress
1. 部署后端服务
# 在相关所有节点下载后端服务镜像,避免镜像下载超时;# 后端服务使用nginx[root@kubenode1 ~]# docker pull nginx # 部署后端服务,这里将后端服务Pod(Deployment的形式下发Pod,直接创建的Pod与Service关联有问题,通过”kubectl get endpoints”可查看到Service的”ENDPOINTS”列关联不上后端Pod,原因未查明)与Service放在1个yaml文件中[root@kubenode1 ~]# cd /usr/local/src/yaml/ingress/[root@kubenode1 ingress]# touch nginx-svc.yaml[root@kubenode1 ingress]# vim nginx-svc.yamlapiVersion: extensions/v1beta1kind: Deploymentmetadata: name: nginx-01spec: replicas: 1 template: metadata: labels: name: nginx-01 spec: containers: - name: nginx-01 image: nginx:latest ports: - containerPort: 80 ---apiVersion: v1kind: Servicemetadata: # Service的全局唯一名称 name: nginx-svcspec: ports: # Service服务监听的端口号 - port: 80 # 后端服务Pod提供的端口号 targetPort: 80 # 端口名称(非必须) name: http # Service关联定义了相应标签的Pod selector: name: nginx-01# 启动后端服务[root@kubenode1 ingress]# kubectl create -f nginx-svc.yaml
2. 部署ingress
[root@kubenode1 ingress]# touch nginx-svc-ingress.yaml [root@kubenode1 ingress]# vim nginx-svc-ingress.yamlapiVersion: extensions/v1beta1kind: Ingressmetadata: name: nginx-svc-ingressspec: rules: # 主机域名,需要在本地绑定节点ip - host: nginx-svc.mehttp: # 如果paths下有具体的路径,如/demo,需要与后端提供真实服务的path一致,这里即nginx下需要有/demo路径 paths: - backend: # 后端服务名 serviceName: nginx-svc # 后端服务监听端口,区别于提供真实服务的容器监听端口 servicePort: 80# 下发ingress[root@kubenode1 ingress]# kubectl create -f nginx-svc-ingress.yaml
3. 验证
# Service关联到后端Pod;# 为主机”nginx-svc.me”定制了ingress策略;# 理论上ingress的”ADDRESS”列显示ingress-nginx-controller Pod的ip地址则表示nginx已设置好后端Service的Endpoint,ingress此时可以正常工作;为空则有需要排错;但可能这里有个bug(在1.8.x与1.9.x版本都有此问题),在显示为空的状态下,ingress依然生效[root@kubenode1 ingress]# cd ~[root@kubenode1 ~]# kubectl get endpoints nginx-svc -o wide[root@kubenode1 ~]# kubectl get ingress -o wide
在本地浏览器访问host主机(注意提前绑定域名):
# 或者采用--resolve参数模拟dns解析,目标地址为域名[root@kubenode1 ~]# curl --resolve nginx-svc.me:80:172.30.200.21 # 或者采用-H参数设置http头中需要访问的域名,目标地址为ip地址[root@kubenode1 ~]# curl -H 'Host:nginx-svc.me'
4. ingress策略配置技巧
1)转发到单个后端服务
# 所有访问被转发到后端唯一的Service,此时可不定义rule# 关注红色加粗字体apiVersion: extensions/v1beta1kind: Ingressmetadata: name: ingressspec: backend: serviceName: nginx-svc servicePort: 80
2)同一域名,不同的url路径被转发到不同的服务
# 相同域名下两个不同的路径对应不同的服务;# 注意如果paths下有具体的路径,如/web,/api等,需要与后端提供真实服务的path一致,这里即nginx服务器下需要有/web,/api等路径apiVersion: extensions/v1beta1kind: Ingressmetadata: name: ingressspec: rules: - host: nginx-svc.me http: paths: - path: /web backend: serviceName: nginx-svc-web servicePort: 80 - path: /api backend: serviceName: nginx-svc-api servicePort: 8081
3)不同域名,被转发到不同的服务
# 不同域名对应不同的服务apiVersion: extensions/v1beta1kind: Ingressmetadata: name: ingressspec: rules: - host: nginx-svc.me http: paths: - backend: serviceName: nginx-svc servicePort: 80 - host: apache-svc.me http: paths: - backend: serviceName: apache-svc servicePort: 80
4)不使用域名转发
# 使用无域名ingress规则时,默认禁用http,强制启用https;# 此时客户端访问如下路径:curl http://172.30.200.21/demo 会返回301错误,但使用https访问则可以成功:curl -k https://172.30.200.21/demo ;# 可以在ingress定义的metadata设置annotation “ingress.kubernetes.io/ssl-redirect=false”关闭强制启用https的设置,如下蓝色加粗字体apiVersion: extensions/v1beta1kind: Ingressmetadata: name: ingress # annotations: # ingress.kubernetes.io/ssl-redirect: “false”spec: rules: - http: paths: - path: /demo backend: serviceName: nginx-svc-demo servicePort: 8080
四.ingress的tls设置
对ingress中的域名进行tls安全证书的设置步骤如下:
- 创建自签名的秘钥与ssl证书;
- 将证书保存到kubernetes集群的1个Secret资源对象上;
- 设置Secret资源对象到ingress中。
根据网站域名是1个还是多个,前两步的操作稍有不同,第3步操作相同,下面以多域名的操作为例:
1. 生成ca证书
[root@kubenode1 ~]# mkdir -p /etc/kubernetes/ingress[root@kubenode1 ~]# cd /etc/kubernetes/ingress/[root@kubenode1 ingress]# openssl genrsa -out ca.key 2048[root@kubenode1 ingress]# openssl req -x509 -new -nodes -key ca.key -days 3560 -out ca.crt -subj "/CN=ingress-ca"
2. 修改openssl.cnf文件
# 对于多域名,生成ssl证书需要使用额外的x509v3配置文件辅助;# 在[alt_names]字段中设置多域名[root@kubenode1 ingress]# cp /etc/pki/tls/openssl.cnf .[root@kubenode1 ingress]# vim openssl.cnf[ req ]# 第126行,取消注释req_extensions = v3_req # The extensions to add to a certificate request[ v3_req ]# Extensions to add to a certificate requestbasicConstraints = CA:FALSEkeyUsage = nonRepudiation, digitalSignature, keyEncipherment# 第224行之后,新增部分subjectAltName = @alt_names[alt_names]DNS.1 = nginx01-svc-tls.meDNS.2 = nginx02-svc-tls.me
3. 生成ingress ssl证书
# 基于修改的openssl.cnf与ca证书生成ingress ssl证书# 生成秘钥[root@kubenode1 ingress]# openssl genrsa -out ingress.key 2048# 生成csr文件[root@kubenode1 ingress]# openssl req -new -key ingress.key -out ingress.csr -subj "/CN=nginx-svc-tls" -config openssl.cnf# 生成证书[root@kubenode1 ingress]# openssl x509 -req -in ingress.csr -CA ca.crt -CAkey ca.key -CAcreateserial -out ingress.crt -days 3650 -extensions v3_req -extfile openssl.cnf
4. 生成Secret资源对象
Secret对象的主要作用是保管私密数据,如:密码,OAuth Tokens,ssh Keys等信息。将私密信息存放在Secret对象中,比直接放在Pod或者Docker image中更安全,更便于使用与分发。
Secret对象创建完成之后,可通过3种方式调用:
- 在创建Pod时,通过为Pod指定Service Account来自动使用;
- 通过挂载Secret到Pod来使用;
- Docker image下载时使用,通过指定Pod的spc.ImagePullSecrets来引用。
# 编辑secret-ingress.yaml文件,将ingress.key与ingress.crt的内容复制到yaml文件中;# 注意1:Secret的”data”域的各子域的值必须为BASE64编码;# 注意2:复制key与crt的内容时去掉换行符,变成一行[root@kubenode1 ingress]# cd /usr/local/src/yaml/ingress/[root@kubenode1 ingress]# touch secret-ingress.yaml[root@kubenode1 ingress]# vim secret-ingress.yamlapiVersion: v1kind: Secretmetadata: name: secret-ingress# 1.8.x之后使用kubernetes.io/tls替换Opaquetype: kubernetes.io/tlsdata: tls.crt: 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 tls.key: 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# 生成Secret资源对象[root@kubenode1 ingress]# kubectl create -f secret-ingress.yaml# 以上编辑yaml文件,使用”kubectl create”命令生成Secret对象在步骤上更清晰;# 但可以利用”kubectl create secret tls”命令直接创建Secret对象[root@kubenode1 ingress]# kubectl create secret tls secret-ingress --key /etc/kubernetes/ingress/ingress.key --cert /etc/kubernetes/ingress/ingress.crt
5. 创建后端服务
# 编辑后端服务nginx01-svc-tls.yaml[root@kubenode1 ingress]# touch nginx01-svc-tls.yaml[root@kubenode1 ingress]# vim nginx01-svc-tls.yamlapiVersion: extensions/v1beta1kind: Deploymentmetadata: name: nginx01-tlsspec: replicas: 1 template: metadata: labels: name: nginx01-tls spec: containers: - name: nginx01-tls image: nginx:latest ports: - containerPort: 80---apiVersion: v1kind: Servicemetadata: name: nginx01-svc-tlsspec: ports: # Service服务监听的端口号 - port: 443 # 后端提供真实服务的Pod提供的端口号 targetPort: 80 name: https selector: name: nginx01-tls# 编辑后端服务nginx02-svc-tls.yaml[root@kubenode1 ingress]# cp nginx01-svc-tls.yaml nginx02-svc-tls.yaml[root@kubenode1 ingress]# sed -i 's|nginx01|nginx02|g' nginx02-svc-tls.yaml# 生成后端服务[root@kubenode1 ingress]# kubectl create -f nginx01-svc-tls.yaml [root@kubenode1 ingress]# kubectl create -f nginx02-svc-tls.yaml
# 修改提供后端服务的nginx容器的html文件;# 通过”kubectl exec -ti-c /bin/bash”进入容器修改;pod-name可通过命令”kubectl get pods -o wide”获取;container-name即yaml文件中定义的名字[root@kubenode1 ingress]# kubectl get pods -o wide
# nginx官方容器的index.html文件在/usr/share/nginx/html/目录下[root@kubenode1 ingress]# kubectl exec -ti nginx01-tls-59fbf6696c-qfq4k -c nginx01-tls /bin/bashroot@nginx01-tls-59fbf6696c-qfq4k:/# echo "Welcome to test site nginx01-svc-tls
" > /usr/share/nginx/html/index.htmlroot@nginx01-tls-59fbf6696c-qfq4k:/# cat /usr/share/nginx/html/index.html root@nginx01-tls-59fbf6696c-qfq4k:/# exit[root@kubenode1 ingress]# kubectl exec -ti nginx02-tls-5559fd9bc7-dfbrp -c nginx02-tls /bin/bashroot@nginx02-tls-5559fd9bc7-dfbrp:/# echo "Welcome to test site nginx02-svc-tls
" > /usr/share/nginx/html/index.htmlroot@nginx02-tls-5559fd9bc7-dfbrp:/# cat /usr/share/nginx/html/index.html root@nginx02-tls-5559fd9bc7-dfbrp:/# exit
6. 创建ingress对象
# 编辑ingress对象yaml文件;# 在”spec”域下新增“tls”子域,”hosts”字段加入多域名,“secretName”字段调用对应的Secret资源;# 1个ingress对象只能使用1个Secret对象(“secretName”字段value唯一),即只能使用1个证书,该正式需要支持”hosts”字段下所有域名;# “secretName”字段一定要置于域名列表最后的位置;# ”hosts”字段的域名需要匹配”rules”字段域名;# ingress默认情况下,当不配置证书或者证书配置错误时,会默认给出一个tls证书;如“secretName”字段配置了2个值,则所有域名采用默认证书;如”hosts”字段少配置一个域名,缺失的域名会采用默认证书;# 更新ingress证书可能需要等待一段时间才能生效[root@kubenode1 ingress]# touch nginx-svc-tls-ingress.yaml[root@kubenode1 ingress]# vim nginx-svc-tls-ingress.yamlapiVersion: extensions/v1beta1kind: Ingressmetadata: name: ingress-tlsspec: tls: - hosts: - nginx01-svc-tls.me - nginx02-svc-tls.me secretName: secret-ingress rules: - host: nginx01-svc-tls.me http: paths: - backend: serviceName: nginx01-svc-tls # 后端服务监听端口,区别于提供真实服务的容器监听端口 servicePort: 443 - host: nginx02-svc-tls.me http: paths: - backend: serviceName: nginx02-svc-tls servicePort: 443# 生成ingress对象[root@kubenode1 ingress]# kubectl create -f nginx-svc-tls-ingress.yaml[root@kubenode1 ingress]# kubectl get ingress
7. 验证
# 采用--resolve参数模拟dns解析,目标地址为域名;# http访问时被重定向,采用https访问正常[root@kubenode1 ingress]# curl --resolve nginx01-svc-tls.me:80:172.30.200.21 [root@kubenode1 ingress]# curl --resolve nginx01-svc-tls.me:443:172.30.200.21 -k
# 或者采用-H参数设置http头中需要访问的域名,目标地址为ip地址[root@kubenode1 ingress]# curl -H 'Host:nginx01-svc-tls.me' -k [root@kubenode1 ingress]# curl -H 'Host:nginx02-svc-tls.me' -k
在本地浏览器访问host主机(注意提前绑定域名):
采用http访问,重定向自动跳转为https访问,如下:
站点:nginx01-svc-tls.me
站点:nginx02-svc-tls.me