I reinstalled a k3s cluster built with --no-extras and added `--k3s-extra-args="--flannel-backend=host-gw". The cluster failed to start.

I logged in and saw that the output of ps showed the following:

14760 ?        Ssl    0:40 /usr/local/bin/k3s server --cluster-init --tls-san 10.68.0.63 --flannel-backend=host-gw--n

Given an existing cluster, is there any way to check
A) whether a local registry is enabled for that cluster, and
B) the value of registry-port/registry-name?

Relatedly: given a cluster that was started without the --enable-registry flag, is there any way to enable the registry post-facto?

In Kubernetes 1.18 the Ingress definitions have changed slightly.

In particular the @paurosello mentioned the change about ingressClass becoming part of the spec, instead of being an annotation.

If we can add both, perhaps that will help keep backwards compatibility for the time being?

https://kubernetes.io/docs/set

Some clusters can be stopped when k3x is started. Users could think the current cluster should be ready, but they would really need to "start" it. We should show a warning saying something like "the current cluster is stopped".

The README.md should be more clear about which host needs the dependency:

The control host requires the following Python dependencies:

    jmespath >= 0.9.0

"The control host" could be easily mistaken with the control plane.