NAT Security rule

[Glitchen]

I'm used to working on Cisco ASA and I'm having a hard time understanding why the security rule states Untrust-L3 for both the source and destination zone. Typically wouldn't that be Untrust-L3 to DMZ? Is there a specific reason for this behavior?

[kiwi]

Hi @Glitchen,

This is by design.

Check out the following articles that can be very helpful on this topic :

https://live.paloaltonetworks.com/t5/Learning-Articles/Packet-Flow-Sequence-in-PAN-OS/ta-p/56081

https://live.paloaltonetworks.com/t5/Tutorials/Getting-Started-Network-Address-Translation-NAT/ta-p/...

Cheers !

-Kiwi.

[reaper · Accepted solution]

Hi @Glitchen

When a new packet arrives on an interface, the firewall needs to determine it's source and destination zones, these are determined by using the routing table.

So when a packet arrives from the internet, both it's source as destination IPs will be located in the untrust zone: they both follow the default gateway when performing a route lookup

(eg. packet from 193.51.100.1 to 203.0.113.1, both ip will resolve via 0.0.0.0/0 to the untrust zone)

Next, the NAT policy is checked to see if this packet matches something, and if it does, which zone the final destination is located in, this is where, post translation, it is deterined that security wise the (final) destination is dmz (eg. 203.0.113.1 translates to 10.0.0.1, which is routed to the dmz interface).

The next step is to verify the security policy for matches to see if the packet is going to be allowed or not. Since we now know the final destination is DMZ, you can build a security policy to allow from untrust to dmz