在前文中介绍了在kubernetes 1.5.2集群环境中使用traefik进行服务发布。Traefik采用daemonset方式部署,连接api-server走的是http协议,也未配置rbac。本文将介绍在k8s 1.9版本中使用deployment方式部署traefik来进行服务发布。
在开始之前,需要先了解一下什么是RBAC。RBAC(基于角色的访问控制)使用 rbac.authorization.k8s.io API 组来实现权限控制,RBAC 允许管理员通过 Kubernetes API 动态的配置权限策略。在 1.6 版本中 RBAC 还处于 Beat 阶段,如果想要开启 RBAC 授权模式需要在 apiserver 组件中指定 --authorization-mode=RBAC 选项。
在 RBAC API 的四个重要概念:
Role:是一系列的权限的集合,例如一个角色可以包含读取 Pod 的权限和列出 Pod 的权限
ClusterRole: 跟 Role 类似,但是可以在集群中到处使用( Role 是 namespace 一级的)
RoloBinding:把角色映射到用户,从而让这些用户继承角色在 namespace 中的权限。
ClusterRoleBinding: 让用户继承 ClusterRole 在整个集群中的权限。
简单点说RBAC实现了在k8s集群中对api-server的鉴权,更多的RBAC知识点请查阅官方文档:https://kubernetes.io/docs/admin/authorization/rbac/
一、给集群的节点打上label
因为选择deployment方式部署,所以要给集群的节点打上label,后续选择nodeSelector指定traefik=proxy,副本数和集群节点数一致的时候,所有的节点上都会运行一个pod
# kubectl get nodes --show-labels
# kubectl label node vm1 traefik=proxy
# kubectl label node vm2 traefik=proxy
# kubectl get nodes --show-labels
# cat traefik-rbac.yaml
---
kind: ClusterRole
apiVersion: rbac.authorization.k8s.io/v1beta1
metadata:
name: traefik-ingress-controller
rules:
- apiGroups:
- ""
resources:
- services
- endpoints
- secrets
verbs:
- get
- list
- watch
- apiGroups:
- extensions
resources:
- ingresses
verbs:
- get
- list
- watch
---
kind: ClusterRoleBinding
apiVersion: rbac.authorization.k8s.io/v1beta1
metadata:
name: traefik-ingress-controller
roleRef:
apiGroup: rbac.authorization.k8s.io
kind: ClusterRole
name: traefik-ingress-controller
subjects:
- kind: ServiceAccount
name: traefik-ingress-controller
namespace: kube-system
在启用rbac的环境下,如果鉴权未配置清楚,则traefik pod会报错如下
E0226 00:15:27.729832 1 reflector.go:199] github.com/containous/traefik/vendor/k8s.io/client-go/tools/cache/reflector.go:94: Failed to list *v1.Service: services is forbidden: User "system:serviceaccount:kube-system:default" cannot list services at the cluster scope
E0226 00:15:29.013298 1 reflector.go:199] github.com/containous/traefik/vendor/k8s.io/client-go/tools/cache/reflector.go:94: Failed to list *v1.Endpoints: endpoints is forbidden: User "system:serviceaccount:kube-system:default" cannot list endpoints at the cluster scope
E0226 00:15:29.213354 1 reflector.go:199] github.com/containous/traefik/vendor/k8s.io/client-go/tools/cache/reflector.go:94: Failed to list *v1.Secret: secrets is forbidden: User "system:serviceaccount:kube-system:default" cannot list secrets at the cluster scope
E0226 00:15:29.698574 1 reflector.go:199] github.com/containous/traefik/vendor/k8s.io/client-go/tools/cache/reflector.go:94: Failed to list *v1beta1.Ingress: ingresses.extensions is forbidden: User "system:serviceaccount:kube-system:default" cannot list ingresses.extensions at the cluster scope
E0226 00:15:30.411837 1 reflector.go:199] github.com/containous/traefik/vendor/k8s.io/client-go/tools/cache/reflector.go:94: Failed to list *v1.Service: services is forbidden: User "system:serviceaccount:kube-system:default" cannot list services at the cluster scope
E0226 00:15:31.912887 1 reflector.go:199] github.com/containous/traefik/vendor/k8s.io/client-go/tools/cache/reflector.go:94: Failed to list *v1.Endpoints: endpoints is forbidden: User "system:serviceaccount:kube-system:default" cannot list endpoints at the cluster scope
2、traefik的deployment文件
# cat traefik-deployment.yaml
---
apiVersion: v1
kind: ServiceAccount
metadata:
name: traefik-ingress-controller
namespace: kube-system
---
kind: Deployment
apiVersion: extensions/v1beta1
metadata:
name: traefik-ingress-controller
namespace: kube-system
labels:
k8s-app: traefik-ingress-lb
spec:
replicas: 2
selector:
matchLabels:
k8s-app: traefik-ingress-lb
template:
metadata:
labels:
k8s-app: traefik-ingress-lb
name: traefik-ingress-lb
spec:
serviceAccountName: traefik-ingress-controller
hostNetwork: true
nodeSelector:
traefik: proxy
terminationGracePeriodSeconds: 60
containers:
- image: traefik
name: traefik-ingress-lb
ports:
- name: web
containerPort: 80
hostPort: 80
- name: admin
containerPort: 8081
args:
- --web
- --web.address=:8081
- --kubernetes
3、traefik的service文件
# cat traefik-service.yaml
apiVersion: v1
kind: Service
metadata:
name: traefik-web-ui
namespace: kube-system
spec:
selector:
k8s-app: traefik-ingress-lb
ports:
- port: 80
targetPort: 8081
4、通过yaml文件创建clusterrole、clusterrolebinding、deployment、serviceaccount、service
# ls
# kubectl create -f traefik-rbac.yaml
# kubectl create -f traefik-deployment.yaml
# kubectl create -f traefik-service.yaml
# kubectl get pod -n kube-system
# kubectl get svc -n kube-system
# kubectl get svc
可以看到集群中default namespace中存在一个frontend服务。kube-system namespace中存在nginx-test、traefik-web-ui、kubernetes-dashboard三个服务。我们后续将创建4个ingress
通过web-ui可以看到在两个节点上各运行了一个pod
三、通过yaml文件创建ingress
# cat ui.yaml
apiVersion: v1
kind: Service
metadata:
name: traefik-web-ui
namespace: kube-system
spec:
selector:
k8s-app: traefik-ingress-lb
ports:
- port: 80
targetPort: 8081
---
apiVersion: extensions/v1beta1
kind: Ingress
metadata:
name: traefik-web-ui
namespace: kube-system
annotations:
kubernetes.io/ingress.class: traefik
spec:
rules:
- host: traefik-ui
http:
paths:
- backend:
serviceName: traefik-web-ui
servicePort: 80
# cat webui-ing.yaml
apiVersion: extensions/v1beta1
kind: Ingress
metadata:
name: traefik-ingress
namespace: kube-system
annotations:
kubernetes.io/ingress.class: traefik
spec:
rules:
- host: k8s.webui
http:
paths:
- backend:
serviceName: kubernetes-dashboard
servicePort: 443
# cat redis-ing.yaml
apiVersion: extensions/v1beta1
kind: Ingress
metadata:
name: traefik-ingress
namespace: default
annotations:
kubernetes.io/ingress.class: traefik
spec:
rules:
- host: k8s.frontend
http:
paths:
- backend:
serviceName: frontend
servicePort: 80
# cat nginx-ing.yaml
apiVersion: extensions/v1beta1
kind: Ingress
metadata:
name: traefik-nginx-ingress
namespace: kube-system
annotations:
kubernetes.io/ingress.class: traefik
spec:
rules:
- host: test.fjhb.cn
http:
paths:
- backend:
serviceName: nginx-test
servicePort: 80
# kubectl create -f ui.yaml
# kubectl create -f webui-ing.yaml
# kubectl create -f redis-ing.yaml
# kubectl get ingress
# kubectl get ingress -n kube-system
三、验证
1、通过访问traefik service对应的nodeport端口,4个ingress配置都加载到了
2、修改测试机hosts文件,将4个域名的解析分配到两台节点上
3、浏览器访问测试
这里出现500错误的原因是,后端的kubernetes-dashboard配置的是https协议
可以在health页面看到http状态码的统计信息