fix #1788 lock dashboard version to 1.6.3 #1805
Conversation
k8s-ci-robot
added
cncf-cla: yes
| dashboard_image_repo: kubernetesdashboarddev/kubernetes-dashboard-amd64 | ||
| dashboard_image_tag: head | ||
| dashboard_image_repo: gcr.io/google_containers/kubernetes-dashboard-amd64 | ||
| dashboard_image_tag: 1.6.3 |
There was a problem hiding this comment.
Please change to v1.6.3 (the whole tag)
| @@ -62,7 +62,7 @@ spec: | |||
| spec: | |||
| containers: | |||
| - name: kubernetes-dashboard | |||
| image: {{ dashboard_image_repo }}:{{ dashboard_image_tag }} | |||
| image: {{ dashboard_image_repo }}:v{{ dashboard_image_tag }} | |||
|
ci check this |
|
Tested locally and it works. |
|
@shadycuz I don't understand your question. Dashboard with basic auth works with this PR. |
|
@mattymo When I asked you in slack you said what broke basic auth was how kubelet mounted the kubernetes folder? So looking at the PR I wasn't sure this actually fixed it and you closed my issues so I wanted to Double check =). My basic AUTH issue that you closed has nothing to do with dashboard and I tried a different 1.6.3 dashboard and it still didn't work. So I don't see how changing the dashboard fixes my auth problem? |
|
@shadycuz that's kubeadm mode. Dashboard auth was broken for both kubeadm and non-kubeadm mode. |
|
Yeah but my basic auth is just broken completely. If you look at my issue you will see using basic auth I'm not even able to list nodes through http post #1803 , I used all defaults when standing up cluster. So if I rerun my playbook basic auth will be fixed or should I stand up a new cluster? @mattymo |
partially fix #1788 lock dashboard version to 1.6.3 while we get 1.7.x working