The purpose of spoof protection is to ensure and enforce that the source address of any packet is in the network which is connected to and reachable through the interface the packet is coming from and only from and through that interface.

However, network topologies can be very complex which renders the spoof protect ruleset very complicated. The ruleset generator automates the generation of these types of rules as well.

To successfully generate a secure ruleset, the generator must be aware of the network topology of the neighbourhood of the firewall. In the Networking component for every interface you have to configure the connected (Connects) zones and set the Spoof protection also.

The connected zones sets which zones, which network addresses are reachable through that interface. The core of the spoof protection is the spoof chain which is in the filter table. The spoof chain is jumped from the INPUT and FORWARD chains as both incoming and forwarded connections must be checked against spoof attacks. The chain begins with the crucial head group followed by the loopback interface handling.

First every packet is ACCEPTed that comes in on the LO interface, then everything is LOGed (with a 'filter/spoof DROP' prefix) and DROPed that has source address in the 127.0.0.0/8 network, but the incoming interface is not LO. Spoof check is done on a per-interface basis. Since all the interfaces are checked separately an SP<interface name> chain is created for each interface.

After the loopback interface check taking place in the spoof chain according to the incoming interface, the evaluation is jumped to the appropriate SP* chain.

In the SP* chains spoof attacks are checked based on what was configured in the Network component. Every SP* chains has almost the same ruleset, the rules differ only in their target part. The chains contain one or two rules (depending on whether it is a LOG + DROP or a RETURN rule) for every addresses of zones that are connected to the given interface. Every rule has a network address of a zone as a source match.

The order of the rules are very important. The rules are sorted by the netmask of the source match in descending order. This way the rules with the highest netmask comes first, which also means that the smallest networks are matched first. It is inevitable to have this order to match the tightest network first, otherwise the tighter rules would never match, because the bigger networks would match previously.

The target of a rule depends on whether the zone containing the address of the match is selected as a connected zone for the interface to which the chain applies. If it is selected, the target is RETURN otherwise there are two rules: a LOG and a DROP one.

This ruleset ensures that if a zone is connected to an interface, the source address of packets coming in on that interface must match any networks of the zone. It also ensures that if the zone, which contains the network the packet is sourced from, is connected to any other interface then the packet is not allowed, it would be LOGged and DROPped.

Zones that have no addresses associated with or not connected to any interface are not used and ignored.