Hey All, I apologize if this is an easy one but I’m beating my head against the desk. I just moved away from esxi and have a bare metal harvester server running v1.4.0. I have a separate rancher deployment that is not hosted on this server. I am trying to enter the rancher’s cluster-registration-url into the harvester settings but it is failing as dns is not finding an entry for the host.
When I ssh into the harvester server and nslookup my rancher deployment it returns the correct IP. But it appears that when I enter that URL into cluster-registration-url it uses rke2-coredns to attempt to resolve, not the server’s dns provider.
Is this expected behavior? If so, what is the best way to add the harvester server into my rancher deployment?
Yes it’s expected behavior as harvester will deploy a fleet agent in its own rke2 cluster to talk to rancher api. When you ssh into harvester you probably have already dns configured (on the machine), but you don’t have same dns in the cluster running harvester.
To make it work (as you guessed) you need to customize harvester rke2 coredns config to point to your dns (resolving the rancher name) to allow pods in harvester rke2 cluster to resolve your domain names