Traefik K8s的安装问题

时间:2018-10-12 15:46:42

标签: kubernetes traefik kubernetes-helm

背景

我正在尝试搭建BareMetal K8s集群,并希望利用Traefik的集群Ingress的众多功能。我已经在MetalLB的前面提供了LoadBalancer IP地址,这对我来说现在不是问题。

信息:

K8s群集版本:1.12

Helm and Tiller版本:v2.11.0

问题:

如果我使用头盔图表安装了Traefik,并且安装了link,但是当我去检查创建的容器的docker日志时,会得到错误提示

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

如果我改走其​​他方法,并尝试使用官方文档手动安装traefik,则至少可以使它正常工作,但是随后我会在

方面遇到错误

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"

因此,要在我的开发人员(最终产品集群)中启动并运行它,我需要付出100%的努力。谁能提供一些帮助和/或指导,以使我朝正确的方向工作?

提前谢谢

2 个答案:

答案 0 :(得分:2)

对于第一次安装(使用Helm),您似乎缺少了RBAC configs

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

对于第二个安装,看来Traefik可能已配置为从监视名称空间和群集中不存在的prometheus-server服务端点中刮取指标。如果您可以共享部署方式,那就太好了。

答案 1 :(得分:0)

感谢您回答里科。

因此,作为后续测试。我回去炸掉了所有东西,并使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

在创建输出的末尾跟随建议的命令

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

因此,我知道由于 MetalLB ,我至少能够获得一个“外部IP地址”,并且作为初始设置的一部分,它应该为我创建了相关的入口规则,但是我我不确定它是否按照我描述的那样正确地执行了操作。

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>

我看到的重要一点是,它实际上并没有获得应有的后端信息(至少是我了解的方式)。

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

然后在我运行时向我显示以下信息:

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

我也刚回去运行了命令     kubectl记录卡其色-鹅-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

应用完所有这些之后,然后尝试通过端口80或443打开一个Web浏览器到外部IP地址,然后得到未找到的可怕页面错误。我知道这里有很多信息可以参考舵设置方法,但是我想为此提供尽可能多的信息。

我将不得不回头并从命令历史记录中解析出我对第二次设置所做的相关命令,并在正确提取所有内容后尝试在此处回复。