Feature Request:
The k8s charm documents that kube-apiserver-extra-args can only be set at charm deployment time. So the following behaviour is observed:
When kube-apiserver-extra-args is updated on a running cluster via:
juju config k8s kube-apiserver-extra-args="default-not-ready-toleration-seconds=60 default-unreachable-toleration-seconds=60"
The charm processes the config-changed event successfully and juju config k8s kube-apiserver-extra-args shows the expected value, but the args are not propagated to the running kube-apiserver. A newly created pod after the config change still shows the default tolerationSeconds: 300.
Are there plans to enable this option on a running cluster?
Feature Request:
The k8s charm documents that
kube-apiserver-extra-argscan only be set at charm deployment time. So the following behaviour is observed:When kube-apiserver-extra-args is updated on a running cluster via:
juju config k8s kube-apiserver-extra-args="default-not-ready-toleration-seconds=60 default-unreachable-toleration-seconds=60"The charm processes the config-changed event successfully and
juju config k8s kube-apiserver-extra-argsshows the expected value, but the args are not propagated to the running kube-apiserver. A newly created pod after the config change still shows the defaulttolerationSeconds: 300.Are there plans to enable this option on a running cluster?