K3s with etcd: nodes cannot connect to first node

Im trying to install the latest k3s with the embedded etcd DB.

On the first node I run:

curl -sfL https://get.k3s.io | INSTALL_K3S_CHANNEL=latest sh -s - --write-kubeconfig-mode 644 --cluster-init --token blah

this runs without an error.

On the second node I do:

curl -sfL https://get.k3s.io | INSTALL_K3S_CHANNEL=latest sh -s - --server https://192.168.0.118:6443 --token blah

this stops at:

[INFO] systemd: Starting k3s

on the first node I get the following errors:

Oct 08 10:01:47 k3s-04 k3s[14754]: failed to reach the peerURL(https://192.168.0.119:2380) of member a34017653b7bf464 (Get "https://192.168.0.119:2380/ve>
Oct 08 10:01:47 k3s-04 k3s[14754]: cannot get the version of member a34017653b7bf464 (Get “https://192.168.0.119:2380/version”: Forbidden)
Oct 08 10:01:50 k3s-04 k3s[14754]: health check for peer a34017653b7bf464 could not connect: Forbidden

If I do a “curl https://localhost:2380/version” on the second node I get a “connection refused”.
The same command with the stable version works.

Does anybody know where I’m stuck here?

Issue was a lowercase no_proxy Variable, set to uppercase NO_PROXY and it works.