Kubectl：带有 minikube 的 Kubernetes 超时答案

【问题标题】：Kubectl: Kubernetes with minikube times outKubectl：带有 minikube 的 Kubernetes 超时
【发布时间】：2016-11-14 22:46:25
【问题描述】：

我已经安装了 minikube 并启动了它内置的 Kubernertes 集群

$ minikube start
Starting local Kubernetes cluster...
Kubernetes is available at https://192.168.99.100:443.
Kubectl is now configured to use the cluster.

我也安装了 kubectl

$ kubectl version
Client Version: version.Info{Major:"1", Minor:"3", GitVersion:"v1.3.0", GitCommit:"283137936a498aed572ee22af6774b6fb6e9fd94", GitTreeState:"clean", BuildDate:"2016-07-01T19:26:38Z", GoVersion:"go1.6.2", Compiler:"gc", Platform:"linux/amd64"}

但我无法成功使用 kubectl 与正在运行的 Kubernetes 集群对话

$ kubectl get nodes
Unable to connect to the server: net/http: TLS handshake timeout

编辑

$ minikube logs 

E0712 19:02:08.767815    1257 docker_manager.go:1955] Failed to create pod infra container: ImagePullBackOff; Skipping pod "kube-addon-manager-minikubevm_kube-system(48abed82af93bb0b941173334110923f)": Back-off pulling image "gcr.io/google_containers/pause-amd64:3.0"
E0712 19:02:08.767875    1257 pod_workers.go:183] Error syncing pod 48abed82af93bb0b941173334110923f, skipping: failed to "StartContainer" for "POD" with ImagePullBackOff: "Back-off pulling image \"gcr.io/google_containers/pause-amd64:3.0\""
E0712 19:02:23.767380    1257 docker_manager.go:1955] Failed to create pod infra container: ImagePullBackOff; Skipping pod "kube-addon-manager-minikubevm_kube-system(48abed82af93bb0b941173334110923f)": Back-off pulling image "gcr.io/google_containers/pause-amd64:3.0"
E0712 19:02:23.767464    1257 pod_workers.go:183] Error syncing pod 48abed82af93bb0b941173334110923f, skipping: failed to "StartContainer" for "POD" with ImagePullBackOff: "Back-off pulling image \"gcr.io/google_containers/pause-amd64:3.0\""
E0712 19:02:36.766696    1257 docker_manager.go:1955] Failed to create pod infra container: ImagePullBackOff; Skipping pod "kube-addon-manager-minikubevm_kube-system(48abed82af93bb0b941173334110923f)": Back-off pulling image "gcr.io/google_containers/pause-amd64:3.0"
E0712 19:02:36.766760    1257 pod_workers.go:183] Error syncing pod 48abed82af93bb0b941173334110923f, skipping: failed to "StartContainer" for "POD" with ImagePullBackOff: "Back-off pulling image \"gcr.io/google_containers/pause-amd64:3.0\""
E0712 19:02:51.767621    1257 docker_manager.go:1955] Failed to create pod infra container: ImagePullBackOff; Skipping pod "kube-addon-manager-minikubevm_kube-system(48abed82af93bb0b941173334110923f)": Back-off pulling image "gcr.io/google_containers/pause-amd64:3.0"
E0712 19:02:51.767672    1257 pod_workers.go:183] Error syncing pod 48abed82af93bb0b941173334110923f, skipping: failed to "StartContainer" for "POD" with ImagePullBackOff: "Back-off pulling image \"gcr.io/google_containers/pause-amd64:3.0\""
E0712 19:03:02.766548    1257 docker_manager.go:1955] Failed to create pod infra container: ImagePullBackOff; Skipping pod "kube-addon-manager-minikubevm_kube-system(48abed82af93bb0b941173334110923f)": Back-off pulling image "gcr.io/google_containers/pause-amd64:3.0"
E0712 19:03:02.766609    1257 pod_workers.go:183] Error syncing pod 48abed82af93bb0b941173334110923f, skipping: failed to "StartContainer" for "POD" with ImagePullBackOff: "Back-off pulling image \"gcr.io/google_containers/pause-amd64:3.0\""
E0712 19:03:16.766831    1257 docker_manager.go:1955] Failed to create pod infra container: ImagePullBackOff; Skipping pod "kube-addon-manager-minikubevm_kube-system(48abed82af93bb0b941173334110923f)": Back-off pulling image "gcr.io/google_containers/pause-amd64:3.0"
E0712 19:03:16.766904    1257 pod_workers.go:183] Error syncing pod 48abed82af93bb0b941173334110923f, skipping: failed to "StartContainer" for "POD" with ImagePullBackOff: "Back-off pulling image \"gcr.io/google_containers/pause-amd64:3.0\""
E0712 19:04:15.829223    1257 docker_manager.go:1955] Failed to create pod infra container: ErrImagePull; Skipping pod "kube-addon-manager-minikubevm_kube-system(48abed82af93bb0b941173334110923f)": image pull failed for gcr.io/google_containers/pause-amd64:3.0, this may be because there are no credentials on this request.  details: (Error response from daemon: Get https://gcr.io/v1/_ping: dial tcp 74.125.28.82:443: i/o timeout)
E0712 19:04:15.829326    1257 pod_workers.go:183] Error syncing pod 48abed82af93bb0b941173334110923f, skipping: failed to "StartContainer" for "POD" with ErrImagePull: "image pull failed for gcr.io/google_containers/pause-amd64:3.0, this may be because there are no credentials on this request.  details: (Error response from daemon: Get https://gcr.io/v1/_ping: dial tcp 74.125.28.82:443: i/o timeout)"
E0712 19:04:31.767536    1257 docker_manager.go:1955] Failed to create pod infra container: ImagePullBackOff; Skipping pod "kube-addon-manager-minikubevm_kube-system(48abed82af93bb0b941173334110923f)": Back-off pulling image "gcr.io/google_containers/pause-amd64:3.0"

【问题讨论】：

minikube logs 有什么有用的吗？
@CJCullen 是的，我刚刚用错误日志更新了问题。
嗯。看起来 docker 无法联系到 gcr.io。你能手动docker pull gcr.io/google_containers/pause-amd64:3.0吗？您是否有任何需要配置的代理设置？（如stackoverflow.com/questions/38301724/…）
@CJCullen 是的，我可以手动 docker pull gcr.io/google_containers/pause-amd64:3.0。在运行 minikube 和 kubectl 之前，我确实有在 bash 提示符下由export http=xxx 配置的代理设置。我是否必须在 minikube 的某些配置文件中指定此设置？看起来。
是的，我猜代理配置必须提供给在 minikube 的虚拟机中运行的 docker 守护进程。我不确定该怎么做...

标签： docker kubernetes kubectl

【解决方案1】：

要让它在代理后运行，您需要进行一些与文档不同的设置。一种。您需要确保与 VM 一起运行的 docker 守护程序可以通过代理访问 Internet。湾。您需要确保在主机上运行的 kubectl 可以在不通过代理的情况下访问虚拟机

使用默认的 kubectl 示例

确保将代理传递到由 minikube 创建的 VM 中（这样可以确保 VM 中的 docker 守护进程可以访问 Internet）

minikube start --vm-driver="kvm" --docker-env="http_proxy=xxx" --docker-env="https_proxy=yyy" start

注意：将 xxx 和 yyy 替换为您的代理设置

获取 VM 在启动时获取的 IP。

minikube ip

注意：每次设置 minikube 时都需要这样做，因为它可以更改

确保 kubectl 无需通过代理即可与此 VM 通信

export no_proxy="127.0.0.1,[minikube_ip]"

现在启动 POD 并对其进行测试

kubectl run hello-minikube --image=gcr.io/google_containers/echoserver:1.4 --port=8080

kubectl expose deployment hello-minikube --type=NodePort

kubectl get pod

curl $(minikube service hello-minikube --url)

【讨论】：

Windows 的情况下为set NO_PROXY="127.0.0.1,[minikube_ip]"
上述解决方案对我不起作用。我正在使用 Windows 8.1 并通过传递代理详细信息来启动 minikube。我还用 minikube ip 设置了 NO_PROXY。仍然收到错误为Unable to connect to the server: net/http: TLS handshake timeout

【解决方案2】：

虽然，这是一个老问题，但在这里回答，以便它可以帮助其他人面临这个问题。

我遇到了错误，因为 minikube VM 分配的内存不足。增加 RAM 应该可以解决这个问题。

Unable to connect to the server: net/http: TLS handshake timeout

minikube VM 默认配置为仅使用 2gb 内存。您可以在 minikube 启动期间使用 --memory 标志更改此值。需要 Minikube 停止、删除并重新启动

【讨论】：

这似乎是一个 github 问题，其中包含如何使用 --memory 标志的示例。 github.com/kubernetes/minikube/issues/567 但是这对我不起作用。 JFYI。

【解决方案3】：

在企业代理后面，设置环境变量如下：

export HTTP_PROXY=http://<proxy hostname:port>
export HTTPS_PROXY=https://<proxy hostname:port>
export NO_PROXY=localhost,127.0.0.1,10.96.0.0/12,192.168.99.0/24,192.168.39.0/24

然后：

minikube start

更多信息here

【讨论】：