Traefik K8s设置问题

Question

背景：

我正试图站出一个BareMetal K8s群集，并希望利用Traefik的群集Ingress的众多功能。 我前面有MetalLB提供LoadBalancer IP地址，这对我来说不是问题。

信息：

K8s群集版本：1.12

Helm和Tiller版本：v2.11.0

问题：

如果我使用舵图和安装链接安装Traefik，但当我去检查docker日志中创建的包含时，我会遇到错误。

 E1012 15:23:50.784829 1 reflector.go:205] github.com/containous/traefik/vendor/k8s.io/client-go/informers/factory.go:86: Failed to list *v1.Endpoints: Unauthorized E1012 15:23:52.279720 1 reflector.go:205] github.com/containous/traefik/vendor/k8s.io/client-go/informers/factory.go:86: Failed to list *v1.Service: Unauthorized E1012 15:23:52.784902 1 reflector.go:205] github.com/containous/traefik/vendor/k8s.io/client-go/informers/factory.go:86: Failed to list *v1beta1.Ingress: Unauthorized 

如果我改为采用不同的路线并尝试使用官方文档手动安装traefik，我至少可以让它有点工作，但我会得到错误的

 time="2018-10-12T12:22:57Z" level=error msg="Service not found for monitoring/prometheus-server" time="2018-10-12T12:22:59Z" level=warning msg="Endpoints not found for monitoring/prometheus-server" 

所以我在我的开发（最终的prod集群）中完成并运行我需要做的事情100％。 任何人都可以提供一些帮助和/或指导，让我朝着正确的方向努力吗？

先感谢您

Answer 1

对于第一次安装（使用Helm），您似乎缺少RBAC配置 ：

kubectl apply -f https://raw.githubusercontent.com/containous/traefik/master/examples/k8s/traefik-rbac.yaml

对于第二次安装，看起来可能会将Traefik配置为从监视命名空间和群集中不存在的prometheus-server服务端点中删除指标。 如果你可以分享你如何部署它会很棒。

Answer 2

谢谢你回答Rico。

所以作为后续测试。 我回去了，把所有东西吹走，并用以下命令重新安装

helm install --values values.yaml stable/traefik

NAME:   khaki-goose
LAST DEPLOYED: Fri Oct 12 14:05:23 2018
NAMESPACE: default
STATUS: DEPLOYED

RESOURCES:
==> v1/Deployment
NAME                     AGE
khaki-goose-traefik  0s

==> v1beta1/Ingress
khaki-goose-traefik-dashboard  0s

==> v1/Pod(related)

NAME                                     READY  STATUS   RESTARTS  AGE
khaki-goose-traefik-dccfdf765-cmfl9  0/1    Pending  0         0s

==> v1/ConfigMap

NAME                     AGE
khaki-goose-traefik  0s

==> v1/Service
khaki-goose-traefik-dashboard  0s
khaki-goose-traefik            0s

在创建输出结束时按建议的命令进行操作

kubectl get svc khaki-goose-traefik --namespace default -w

NAME                  TYPE           CLUSTER-IP      EXTERNAL-IP   PORT(S)                      AGE
khaki-goose-traefik   LoadBalancer   10.109.30.181   172.16.15.1   80:31644/TCP,443:31004/TCP   12s

所以我知道我至少能够得到一个“外部IP地址”，感谢MetalLB，并且作为初始设置的一部分，它应该为我创建相关的入口规则，但我不确定它是否正确，因为我描述它我得到以下。

Name:             khaki-goose-traefik-dashboard
Namespace:        default
Address: 
Default backend:  default-http-backend:80 (<none>)
Rules:
Host                 Path  Backends
----                 ----  --------
traefik-ui.minikube
                      khaki-goose-traefik-dashboard:80 (<none>)
Annotations:
Events:  <none>

我能看到的重要一点是它实际上并没有得到关于后端的信息（至少我理解它的方式）。

然后，我使用官方文档中指定的默认值运行您建议的命令。

kubectl apply -f traefik-rbac.yaml

当我跑步时，它向我显示以下信息：

kubectl describe pods 

Name:               khaki-goose-traefik-dccfdf765-k9pxh
Namespace:          default
Priority:           0
PriorityClassName:  <none>
Node:               jbv-dev-k8wrkr-02.removed.domain.name/172.16.1.47
Start Time:         Fri, 12 Oct 2018 14:54:30 -0400
Labels:             app=traefik
                    chart=traefik-1.43.0
                    heritage=Tiller
                    pod-template-hash=dccfdf765
                    release=khaki-goose
Annotations:        checksum/config=fba7308c36f00b969971dec5caa6ca1e456737d4564d6c623d261aee9ffb82cc
Status:             Running
IP:                 172.25.4.44
Controlled By:      ReplicaSet/khaki-goose-traefik-dccfdf765
Containers:
  khaki-goose-traefik:
    Container ID:  docker://f88d222ccf72c244bdb100b956a8b629bcb8b89c9954de1f83552bddd6c44a43
    Image:         traefik:1.6.6
    Image ID:      docker-pullable://traefik@sha256:9569c56e8b7353c9c4e5d4f00177b0b7c523db6926a42a148e04a6fa4b6f9f8d
    Ports:         80/TCP, 8880/TCP, 443/TCP, 8080/TCP
    Host Ports:    0/TCP, 0/TCP, 0/TCP, 0/TCP
    Args:
      --configfile=/config/traefik.toml
    State:          Running
      Started:      Fri, 12 Oct 2018 14:54:41 -0400
    Ready:          True
    Restart Count:  0
    Limits:
      cpu:     100m
      memory:  30Mi
    Requests:
      cpu:        100m
      memory:     20Mi
    Liveness:     tcp-socket :80 delay=10s timeout=2s period=10s #success=1 #failure=3
    Readiness:    tcp-socket :80 delay=10s timeout=2s period=10s #success=1 #failure=1
    Environment:  <none>
    Mounts:
      /config from config (rw)
      /var/run/secrets/kubernetes.io/serviceaccount from default-token-n8865 (ro)
Conditions:
  Type              Status
  Initialized       True
  Ready             True
  ContainersReady   True
  PodScheduled      True
Volumes:
  config:
    Type:      ConfigMap (a volume populated by a ConfigMap)
    Name:      khaki-goose-traefik
    Optional:  false
  default-token-n8865:
    Type:        Secret (a volume populated by a Secret)
    SecretName:  default-token-n8865
    Optional:    false
QoS Class:       Burstable
Node-Selectors:  <none>
Tolerations:     node.kubernetes.io/not-ready:NoExecute for 300s
                 node.kubernetes.io/unreachable:NoExecute for 300s
Events:
  Type    Reason     Age   From                                            Message
  ----    ------     ----  ----                                            -------
  Normal  Scheduled  1m    default-scheduler                               Successfully assigned default/khaki-goose-traefik-dccfdf765-k9pxh to jbv-dev-k8wrkr-02.removed.domain.name
  Normal  Pulled     1m    kubelet, jbv-dev-k8wrkr-02.removed.domain.name  Container image "traefik:1.6.6" already present on machine
  Normal  Created    1m    kubelet, jbv-dev-k8wrkr-02.removed.domain.name  Created container
  Normal  Started    1m    kubelet, jbv-dev-k8wrkr-02.removed.domain.name  Started container

我也回去跑了命令kubectl logs khaki-goose-traefik-dccfdf765-k9pxh

E1012 20:19:14.410278       1 reflector.go:205] github.com/containous/traefik/vendor/k8s.io/client-go/informers/factory.go:86: Failed to list *v1beta1.Ingress: Unauthorized
E1012 20:19:14.411657       1 reflector.go:205] github.com/containous/traefik/vendor/k8s.io/client-go/informers/factory.go:86: Failed to list *v1.Endpoints: Unauthorized
E1012 20:19:15.388608       1 reflector.go:205] github.com/containous/traefik/vendor/k8s.io/client-go/informers/factory.go:86: Failed to list *v1.Service: Unauthorized
E1012 20:19:15.391057       1 reflector.go:205] github.com/containous/traefik/vendor/k8s.io/client-go/informers/factory.go:86: Failed to list *v1beta1.Ingress: Unauthorized
E1012 20:19:15.397616       1 reflector.go:205] github.com/containous/traefik/vendor/k8s.io/client-go/informers/factory.go:86: Failed to list *v1.Endpoints: Unauthorized
E1012 20:19:15.411714       1 reflector.go:205] github.com/containous/traefik/vendor/k8s.io/client-go/informers/factory.go:86: Failed to list *v1.Service: Unauthorized
E1012 20:19:15.413496       1 reflector.go:205] github.com/containous/traefik/vendor/k8s.io/client-go/informers/factory.go:86: Failed to list *v1beta1.Ingress: Unauthorized
E1012 20:19:15.414809       1 reflector.go:205] github.com/containous/traefik/vendor/k8s.io/client-go/informers/factory.go:86: Failed to list *v1.Endpoints: Unauthorized

在我应用了所有这些之后，我尝试通过端口80或443打开外部IP地址的浏览器，然后找到可怕的页面找不到错误。 我知道这里有很多信息可以参考helm设置方式，但我想为此提供尽可能多的信息。

我将不得不回顾并解析我在命令历史记录中为第二次设置所做的相关命令，并在我正确拉出所有内容时尝试回复此处。