Windows Pod 上的 Kubernetes Minikube 拉取镜像失败
Kubernetes Minikube on Windows Pod failed to pull image
我正在尝试使用 Minikube 和 Hyper-V 在 Windows 上 运行 Kubernetes。我已经成功地 运行 Minikube 使用 minikube start --vm-driver=hyperv --hyperv-virtual-switch=KuberNAT 并检查 minikube status 给我
PS > minikube status
minikube: Running
cluster: Running
kubectl: Correctly Configured: pointing to minikube-vm at 192.168.1.74
但现在我正在尝试使用 kubectl 运行 kubernetes 中的图像，我已经设法让我的 powershell window 指向带有 minikube docker-env | Invoke-Expression 的 kubernetes 虚拟机（仅限 powershell，我无法在命令提示符下执行类似的操作），我必须在每个 powershell 中插入此命令 window 我想用来将图像推送到 kubernetes 图像。
我遇到的问题是我不能 运行 一个容器，我可以 "deploy" 一个带有 kubectl run cloudconfig --image=cloudconfig 的图像，但是创建的 pod 给我这个错误:
Failed to pull image "cloudconfig": rpc error: code = Unknown desc =
Error response from daemon: repository cloudconfig not found: does not
exist or no pull access
如果我运行docker image ls我得到
REPOSITORY TAG IMAGE ID CREATED SIZE
cloudconfig latest 9199d500e746 2 minutes ago 105MB
openjdk 8-jre-alpine 5699ac7295f9 6 days ago 81.4MB
gcr.io/google_containers/k8s-dns-sidecar-amd64 1.14.5 fed89e8b4248 5 weeks ago 41.8MB
gcr.io/google_containers/k8s-dns-kube-dns-amd64 1.14.5 512cd7425a73 5 weeks ago 49.4MB
gcr.io/google_containers/k8s-dns-dnsmasq-nanny-amd64 1.14.5 459944ce8cc4 5 weeks ago 41.4MB
gcr.io/google_containers/kubernetes-dashboard-amd64 v1.7.0 284ec2f8ed6c 5 weeks ago 128MB
gcr.io/google-containers/kube-addon-manager v6.4-beta.2 0a951668696f 4 months ago 79.2MB
gcr.io/google_containers/pause-amd64 3.0 99e59f495ffa 18 months ago 747kB
和docker container ls给我
CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES
3d79bab2a212 gcr.io/google_containers/pause-amd64:3.0 "/pause" 41 seconds ago Up 40 seconds k8s_POD_cloudconfig-88c867589-qpqph_default_ac2dd8bb-bee1-11e7-8e51-00155d00ba16_0
e6723a726c26 gcr.io/google_containers/k8s-dns-sidecar-amd64 "/sidecar --v=2 --..." 43 minutes ago Up 43 minutes k8s_sidecar_kube-dns-6fc954457d-gvmz6_kube-system_f5de0b0e-be4b-11e7-8699-00155d00ba16_1
36980feeff76 gcr.io/google_containers/k8s-dns-dnsmasq-nanny-amd64 "/dnsmasq-nanny -v..." 43 minutes ago Up 43 minutes k8s_dnsmasq_kube-dns-6fc954457d-gvmz6_kube-system_f5de0b0e-be4b-11e7-8699-00155d00ba16_1
d3d61e3861f0 gcr.io/google_containers/k8s-dns-kube-dns-amd64 "/kube-dns --domai..." 43 minutes ago Up 43 minutes k8s_kubedns_kube-dns-6fc954457d-gvmz6_kube-system_f5de0b0e-be4b-11e7-8699-00155d00ba16_1
45208e98bd7f gcr.io/google_containers/kubernetes-dashboard-amd64 "/dashboard --inse..." 43 minutes ago Up 43 minutes k8s_kubernetes-dashboard_kubernetes-dashboard-8hw2w_kube-system_f5bb8241-be4b-11e7-8699-00155d00ba16_1
1fbb3688711d gcr.io/google_containers/pause-amd64:3.0 "/pause" 43 minutes ago Up 43 minutes k8s_POD_kube-dns-6fc954457d-gvmz6_kube-system_f5de0b0e-be4b-11e7-8699-00155d00ba16_1
c71ab868584c gcr.io/google-containers/kube-addon-manager "/opt/kube-addons.sh" 43 minutes ago Up 43 minutes k8s_kube-addon-manager_kube-addon-manager-minikube_kube-system_9831e93c3188555873fdb49f43198eef_1
1df643d195d2 gcr.io/google_containers/pause-amd64:3.0 "/pause" 43 minutes ago Up 43 minutes k8s_POD_kubernetes-dashboard-8hw2w_kube-system_f5bb8241-be4b-11e7-8699-00155d00ba16_1
28f48e09fd46 gcr.io/google_containers/pause-amd64:3.0 "/pause" 43 minutes ago Up 43 minutes k8s_POD_kube-addon-manager-minikube_kube-system_9831e93c3188555873fdb49f43198eef_1
所以我假设所有其他事情都正常工作，而 kubernetes 只是不知何故不知道在哪里可以找到我的图像。有谁知道我如何让它工作？
我找到了解决方案，它是由 this github 线程
中的用户建议的
If anyone else ends up on this thread, the solution that worked for me
was updating the image pull policy, you can find info on this here.
From the docs: Be default, the kubelet will try to pull each image
from the specified registry. You need to update this so it can look
locally.
If you're running from the CLI, add --image-pull-policy=IfNotPresent
to your kubectl run, i.e.
kubectl run some-node-proj --image=my-awesome-local-image:v1
--image-pull-policy=IfNotPresent
添加 --image-pull-policy=IfNotPresent 让我 运行 容器没问题。
或者，我通过将驱动程序从 hyperkit 更改为 docker 来让我的工作。似乎 hyperkit 有网络问题。
minikube start --driver=docker
我正在尝试使用 Minikube 和 Hyper-V 在 Windows 上 运行 Kubernetes。我已经成功地 运行 Minikube 使用 minikube start --vm-driver=hyperv --hyperv-virtual-switch=KuberNAT 并检查 minikube status 给我
PS > minikube status
minikube: Running
cluster: Running
kubectl: Correctly Configured: pointing to minikube-vm at 192.168.1.74
但现在我正在尝试使用 kubectl 运行 kubernetes 中的图像，我已经设法让我的 powershell window 指向带有 minikube docker-env | Invoke-Expression 的 kubernetes 虚拟机（仅限 powershell，我无法在命令提示符下执行类似的操作），我必须在每个 powershell 中插入此命令 window 我想用来将图像推送到 kubernetes 图像。
我遇到的问题是我不能 运行 一个容器，我可以 "deploy" 一个带有 kubectl run cloudconfig --image=cloudconfig 的图像，但是创建的 pod 给我这个错误:
Failed to pull image "cloudconfig": rpc error: code = Unknown desc = Error response from daemon: repository cloudconfig not found: does not exist or no pull access
如果我运行docker image ls我得到
REPOSITORY TAG IMAGE ID CREATED SIZE
cloudconfig latest 9199d500e746 2 minutes ago 105MB
openjdk 8-jre-alpine 5699ac7295f9 6 days ago 81.4MB
gcr.io/google_containers/k8s-dns-sidecar-amd64 1.14.5 fed89e8b4248 5 weeks ago 41.8MB
gcr.io/google_containers/k8s-dns-kube-dns-amd64 1.14.5 512cd7425a73 5 weeks ago 49.4MB
gcr.io/google_containers/k8s-dns-dnsmasq-nanny-amd64 1.14.5 459944ce8cc4 5 weeks ago 41.4MB
gcr.io/google_containers/kubernetes-dashboard-amd64 v1.7.0 284ec2f8ed6c 5 weeks ago 128MB
gcr.io/google-containers/kube-addon-manager v6.4-beta.2 0a951668696f 4 months ago 79.2MB
gcr.io/google_containers/pause-amd64 3.0 99e59f495ffa 18 months ago 747kB
和docker container ls给我
CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES
3d79bab2a212 gcr.io/google_containers/pause-amd64:3.0 "/pause" 41 seconds ago Up 40 seconds k8s_POD_cloudconfig-88c867589-qpqph_default_ac2dd8bb-bee1-11e7-8e51-00155d00ba16_0
e6723a726c26 gcr.io/google_containers/k8s-dns-sidecar-amd64 "/sidecar --v=2 --..." 43 minutes ago Up 43 minutes k8s_sidecar_kube-dns-6fc954457d-gvmz6_kube-system_f5de0b0e-be4b-11e7-8699-00155d00ba16_1
36980feeff76 gcr.io/google_containers/k8s-dns-dnsmasq-nanny-amd64 "/dnsmasq-nanny -v..." 43 minutes ago Up 43 minutes k8s_dnsmasq_kube-dns-6fc954457d-gvmz6_kube-system_f5de0b0e-be4b-11e7-8699-00155d00ba16_1
d3d61e3861f0 gcr.io/google_containers/k8s-dns-kube-dns-amd64 "/kube-dns --domai..." 43 minutes ago Up 43 minutes k8s_kubedns_kube-dns-6fc954457d-gvmz6_kube-system_f5de0b0e-be4b-11e7-8699-00155d00ba16_1
45208e98bd7f gcr.io/google_containers/kubernetes-dashboard-amd64 "/dashboard --inse..." 43 minutes ago Up 43 minutes k8s_kubernetes-dashboard_kubernetes-dashboard-8hw2w_kube-system_f5bb8241-be4b-11e7-8699-00155d00ba16_1
1fbb3688711d gcr.io/google_containers/pause-amd64:3.0 "/pause" 43 minutes ago Up 43 minutes k8s_POD_kube-dns-6fc954457d-gvmz6_kube-system_f5de0b0e-be4b-11e7-8699-00155d00ba16_1
c71ab868584c gcr.io/google-containers/kube-addon-manager "/opt/kube-addons.sh" 43 minutes ago Up 43 minutes k8s_kube-addon-manager_kube-addon-manager-minikube_kube-system_9831e93c3188555873fdb49f43198eef_1
1df643d195d2 gcr.io/google_containers/pause-amd64:3.0 "/pause" 43 minutes ago Up 43 minutes k8s_POD_kubernetes-dashboard-8hw2w_kube-system_f5bb8241-be4b-11e7-8699-00155d00ba16_1
28f48e09fd46 gcr.io/google_containers/pause-amd64:3.0 "/pause" 43 minutes ago Up 43 minutes k8s_POD_kube-addon-manager-minikube_kube-system_9831e93c3188555873fdb49f43198eef_1
所以我假设所有其他事情都正常工作，而 kubernetes 只是不知何故不知道在哪里可以找到我的图像。有谁知道我如何让它工作？
我找到了解决方案，它是由 this github 线程
中的用户建议的If anyone else ends up on this thread, the solution that worked for me was updating the image pull policy, you can find info on this here. From the docs: Be default, the kubelet will try to pull each image from the specified registry. You need to update this so it can look locally.
If you're running from the CLI, add --image-pull-policy=IfNotPresent to your kubectl run, i.e.
kubectl run some-node-proj --image=my-awesome-local-image:v1 --image-pull-policy=IfNotPresent
添加 --image-pull-policy=IfNotPresent 让我 运行 容器没问题。
或者，我通过将驱动程序从 hyperkit 更改为 docker 来让我的工作。似乎 hyperkit 有网络问题。
minikube start --driver=docker