This message was deleted.

try to check the log of POD

Rancher-...

,

Harvester...

in harvester side

I checked the agent POD logs and see that the harvester node is trying to connect to a previous instance of my rancher.

CATTLE_SERVER=<https://rancher.senode.dev> CATTLE_SERVER_VERSION=v2.6.6

The above is pointing to a rancher instance that does not exist. I have updated the harvester setting "cluster-registration-url" to my new instance https://rancher75182.senode.dev.... but seems to not work. Is there a way to force to reset which or update the server URL?

try to add this Rancher URL to your DNS server, or give the IP to harvester

The rancher URL is in my DNS and from the harvester server it does resolve. It's just that somehow the agent is not picking up the new URL

When Harvester installation, there is a step to fill in DNS server, you may add all of the public and local DNS server then.

The issue is not with DNS setup. The agent is pointing to an incorrect URL which no longer exists. How does the agent get the rancher URL?

Cold you try to

use default value

in

harvester

setting

cluster-registration-url

first, save to set it to empty value first; then try to use the new `Rancher registration url`` again.

For some reason it's not letting me click "Use Default Value" button. Which explains why the new URL is not getting used.

cluster-registration-url Active
Age: 155 days

@bland-farmer-13503 @faint-football-12756 Do you know any known issue about this? thanks.

Unfortunately my previous instance of rancher was removed so I didn't get harvester into the new rancher and cannot delete. To save time I have decided to start again by reinstalling harvester. Thanks for your help.