This message was deleted.

never mind, it's something internal. domain controller crashed and dhcp exclusions were not copied to new DC. Now the local rancher cluster nodes have different ip's