@riften said in pfblocker and AD DNS:
I've got two Server 2016 VMs running a domain, both with DNS and DHCP on both. No DHCP on PFSense, only DNS Resolver configured. The IPs listed in PFSense in SYSTEM/GENERAL are all my chosen Internet resolvers (not my internal DNS, they are DNS over TLS Internet resolvers). All my clients have as their DNS, my two DNS servers only, and they get this from the DCs DHCP server and scope options. The DCs list each other as their DNS, and have the PFSense box as their forwarder (not conditional, just regular forwarder). If they can't resolve the request, they use the forwarder so port 53 TCP and UDP are allowed via a lan rule applied to an alias for both my DCs, to the lan interface on PFSENSE. All clients port 53 are blocked in PFSense on a rule below that allow rule, just to make sure they don't get directly out for any reason (say if they get infected with something and their DNS is hack-configured to something malicious). Once PFSense gets the request from my DCs, it then uses the configured DNS servers on the general tab, with the DNS over TLS settings set on the DNS Resolver section.
And what does that have to do with the original question/problem? That's your workflow, OK. But that has potential problems / oversights as well. But besides, I don't get what you wanted to say/add to the topic with telling your setup.
As an additional thought: only blocking udp/tcp53 isn't enough anymore. There are DoT resolvers for client OS' as well that could be used and with Windows (or applications) adding DoH support, that "Pandora's Box" will soon bring fun to all admins debugging DNS failings as well ;)