Question

我也将其发布在serverfault上，但希望可以在这里获得更多的观看次数/反馈：

尝试使用kubeadm进行远程访问，以使kubectl proxy集群中的Dashboard UI正常工作。

Error: 'dial tcp 192.168.2.3:8443: connect: connection refused'
Trying to reach: 'https://192.168.2.3:8443/'

通过远程浏览器访问http://localhost:8001/api/v1/namespaces/kube-system/services/https:kubernetes-dashboard:/proxy/时。

查看API日志，发现出现以下错误：

I1215 20:18:46.601151       1 log.go:172] http: TLS handshake error from 10.21.72.28:50268: remote error: tls: unknown certificate authority
I1215 20:19:15.444580       1 log.go:172] http: TLS handshake error from 10.21.72.28:50271: remote error: tls: unknown certificate authority
I1215 20:19:31.850501       1 log.go:172] http: TLS handshake error from 10.21.72.28:50275: remote error: tls: unknown certificate authority
I1215 20:55:55.574729       1 log.go:172] http: TLS handshake error from 10.21.72.28:50860: remote error: tls: unknown certificate authority
E1215 21:19:47.246642       1 watch.go:233] unable to encode watch object *v1.WatchEvent: write tcp 134.84.53.162:6443->134.84.53.163:38894: write: connection timed out (&streaming.encoder{writer:(*metrics.fancyResponseWriterDelegator)(0xc42d6fecb0), encoder:(*versioning.codec)(0xc429276990), buf:(*bytes.Buffer)(0xc42cae68c0)})

我认为这与无法使Dashboard正常工作有关，如果可以的话，我想知道API服务器的问题是什么。群集中的所有其他东西似乎都在工作。

注意，我有admin.conf在本地运行，并且能够通过kubectl毫无问题地访问集群。

另外，值得注意的是，当我第一次建立集群时，它一直在工作。但是，我遇到了网络问题，必须应用此问题才能使CoreDNS工作Coredns service do not work,but endpoint is ok the other SVCs are normal only except dns，所以我想知道这是否会破坏代理服务？

*编辑*

以下是仪表板窗格的输出：

[gms@thalia0 ~]$ kubectl describe pod kubernetes-dashboard-77fd78f978-tjzxt --namespace=kube-system
Name:               kubernetes-dashboard-77fd78f978-tjzxt
Namespace:          kube-system
Priority:           0
PriorityClassName:  <none>
Node:               thalia2.hostdoman/hostip<redacted>
Start Time:         Sat, 15 Dec 2018 15:17:57 -0600
Labels:             k8s-app=kubernetes-dashboard
                    pod-template-hash=77fd78f978
Annotations:        cni.projectcalico.org/podIP: 192.168.2.3/32
Status:             Running
IP:                 192.168.2.3
Controlled By:      ReplicaSet/kubernetes-dashboard-77fd78f978
Containers:
  kubernetes-dashboard:
    Container ID:  docker://ed5ff580fb7d7b649d2bd1734e5fd80f97c80dec5c8e3b2808d33b8f92e7b472
    Image:         k8s.gcr.io/kubernetes-dashboard-amd64:v1.10.0
    Image ID:      docker-pullable://k8s.gcr.io/kubernetes-dashboard-amd64@sha256:1d2e1229a918f4bc38b5a3f9f5f11302b3e71f8397b492afac7f273a0008776a
    Port:          8443/TCP
    Host Port:     0/TCP
    Args:
      --auto-generate-certificates
    State:          Running
      Started:      Sat, 15 Dec 2018 15:18:04 -0600
    Ready:          True
    Restart Count:  0
    Liveness:       http-get https://:8443/ delay=30s timeout=30s period=10s #success=1 #failure=3
    Environment:    <none>
    Mounts:
      /certs from kubernetes-dashboard-certs (rw)
      /tmp from tmp-volume (rw)
      /var/run/secrets/kubernetes.io/serviceaccount from kubernetes-dashboard-token-mrd9k (ro)
Conditions:
  Type              Status
  Initialized       True
  Ready             True
  ContainersReady   True
  PodScheduled      True
Volumes:
  kubernetes-dashboard-certs:
    Type:        Secret (a volume populated by a Secret)
    SecretName:  kubernetes-dashboard-certs
    Optional:    false
  tmp-volume:
    Type:    EmptyDir (a temporary directory that shares a pod's lifetime)
    Medium:
  kubernetes-dashboard-token-mrd9k:
    Type:        Secret (a volume populated by a Secret)
    SecretName:  kubernetes-dashboard-token-mrd9k
    Optional:    false
QoS Class:       BestEffort
Node-Selectors:  <none>
Tolerations:     node-role.kubernetes.io/master:NoSchedule
                 node.kubernetes.io/not-ready:NoExecute for 300s
                 node.kubernetes.io/unreachable:NoExecute for 300s
Events:          <none>

我检查了服务：

[gms@thalia0 ~]$ kubectl -n kube-system get service kubernetes-dashboard
NAME                   TYPE        CLUSTER-IP     EXTERNAL-IP   PORT(S)   AGE
kubernetes-dashboard   ClusterIP   10.103.93.93   <none>        443/TCP   4d23h

还要注意，如果我从主节点curl http://localhost:8001/api收到了有效的响应。

因此，总而言之，我不确定这些错误中有哪些是无法访问仪表板的根源。

我刚刚将集群升级到1.13.1，希望可以解决此问题，但是，no。

Answer 1

执行kubectl proxy时，只能从本地主机访问默认端口8001。如果将ssh连接到安装了kubernetes的机器，则必须将此端口映射到笔记本电脑或用于ssh的任何设备。

您可以通过以下方式通过ssh来主节点并将8001端口映射到您的本地盒：

ssh -L 8001:localhost:8001 hostname@master_node_IP

Answer 2

我已将群集中的所有节点升级到1.13.1版和voila，仪表板现在可以正常工作了，到目前为止，我还没有应用上面提到的CoreDNS修复程序。

Kubernetes api /仪表板问题

2 个答案: