Fix propagation of proxy settings from kubeone.yaml to apiserver env #1749
Labels
kind/bug
Categorizes issue or PR as related to a bug.
priority/normal
Not that urgent, but is important
sig/cluster-management
Denotes a PR or issue as being assigned to SIG Cluster Management.
Milestone
What happened:
if you are having a definition of proxy and no_proxy settings in kubeone, these are propagated on first run to the apiserver manifest on control plane hosts.
Anyway if you update these proxy values, these are not refreshed after running
kubeone apply
orkubeone apply --force-upgrade
. Only option to trigger that is to change for example the kubernetes version.Example files:
Login is happening here - https://github.com/kubernetes/kubernetes/blob/master/cmd/kubeadm/app/util/env.go#L27
There is also a related cause from k8s - kubernetes/kubernetes@1eeb3f1
What is the expected behavior:
Proxy setting is updated at least when
--force-upgrade
is used.How to reproduce the issue:
Follow the kubeone deployment with proxy settings, then add some new value and run kubeone apply (--force-upgrade) and check that values were not updated on control-plane.
Anything else we need to know?
Information about the environment:
KubeOne version (
kubeone version
): 1.3.3Operating system: CentOS
Provider you're deploying cluster on: bare-metal (static kubeone setup)
Operating system you're deploying on:
The text was updated successfully, but these errors were encountered: