Core dns, calico issue because of variable hostname

HI,
im trying to set up a 2 machine cluster on microk8s 1.23.17.
Unfortunately v1.23 was the only version that was not crashing on me in the installation process.
I want to use gitea, so i need the dns addon, but then dns container in the kube-system namespace as well as the calico containers where all crashing on me.
I also noticed that microk8s where creating new nodes every boot and thats why machine was almost running out of resources. i thought it was a standard behaviour by design, just like old replica sets keep existing in your cluster by default.
But manjax from the docker discord told me that its not by design, so i created a static hostname for my machines, recreated the cluster and now its all working fine. and now there are no new nodes that get created every boot.
cant you at least give hint in the instructions, to set a static hostname?
otherwise i would consider this a bug, since this was killing my machine because oif the redundant nodes
and it makes it reeeeeally hard for a beginner to debugg such issues.
Best regards:
ji