[Previous: Queueing] [Contents] [Next: Traffic Redirection]

PF: Network Address Translation (NAT)

Table of Contents

• Introduction
• How NAT Works
• Enabling NAT
• Configuring NAT
• Bidirectional Mapping (1:1 mapping)
• Translation Rule Exceptions
• Checking NAT Status

Introduction

NAT allows you to take advantage of the reserved address blocks described in RFC 1918. Typically, your internal network will be setup to use one or more of these network blocks. They are:

	10.0.0.0/8       (10.0.0.0 - 10.255.255.255)
	172.16.0.0/12    (172.16.0.0 - 172.31.255.255)
	192.168.0.0/16   (192.168.0.0 - 192.168.255.255)

An OpenBSD system doing NAT will have at least two network adapters, one to the Internet, the other to your internal network. NAT will be translating requests from the internal network so they appear to all be coming from your OpenBSD NAT system.

How NAT Works

When a client on the internal network contacts a machine on the Internet, it sends out IP packets destined for that machine. These packets contain all the addressing information necessary to get them to their destination. NAT is concerned with these pieces of information:

• Source IP address (for example, 192.168.1.35)
• Source TCP or UDP port (for example, 2132)

When the packets pass through the NAT gateway they will be modified so that they appear to be coming from the NAT gateway itself. The NAT gateway will record the changes it makes in its state table so that it can a) reverse the changes on return packets and b) ensure that return packets are passed through the firewall and are not blocked. For example, the following changes might be made:

• Source IP: replaced with the external address of the gateway (for example, 24.5.0.5)
• Source port: replaced with a randomly chosen, unused port on the gateway (for example, 53136)

Neither the internal machine nor the Internet host is aware of these translation steps. To the internal machine, the NAT system is simply an Internet gateway. To the Internet host, the packets appear to come directly from the NAT system; it is completely unaware that the internal workstation even exists.

When the Internet host replies to the internal machine's packets, they will be addressed to the NAT gateway's external IP (24.5.0.5) at the translation port (53136). The NAT gateway will then search the state table to determine if the reply packets match an already established connection. A unique match will be found based on the IP/port combination which tells PF the packets belong to a connection initiated by the internal machine 192.168.1.35. PF will then make the opposite changes it made to the outgoing packets and forward the reply packets on to the internal machine.

Translation of ICMP packets happens in a similar fashion but without the source port modification.

Enabling NAT

To enable NAT on an OpenBSD gateway, in addition to activating PF, you must also enable IP forwarding:

# sysctl -w net.inet.ip.forwarding=1
# sysctl -w net.inet6.ip6.forwarding=1 (if using IPv6)

To make this change permanent, the following lines should be added to /etc/sysctl.conf:

net.inet.ip.forwarding=1
net.inet6.ip6.forwarding=1

These lines are present but commented out (prefixed with a #) in the default install. Remove the # and save the file. IP forwarding will be enabled when the machine is rebooted.

Configuring NAT

nat on extif [af] from src_addr [port src_port] to \
   dst_addr [port dst_port] -> ext_addr

extif

The name of the external network interface.

af

The address family, either inet for IPv4 or inet6 for IPv6. PF is usually able to determine this parameter based on the source/destination address(es).

src_addr

The source (internal) address of packets that will be translated. The source address can be specified as:

• A single IPv4 or IPv6 address.
• A CIDR network block.
• A fully qualified domain name that will be resolved via DNS when the ruleset is loaded. All resulting IP addresses will be substituted into the rule.
• The name of a network interface. Any IP addresses assigned to the interface will be substituted into the rule at load time.
• The name of a network interface followed by /netmask (e.g. /24). Each IP address on the interface is combined with the netmask to form a CIDR network block which is substituted into the rule.
• The name of a network interface followed by the :network keyword. The resulting CIDR network (e.g. 192.168.0.0/24) will be substituted into the rule when the ruleset is loaded.
• A table.
• Any of the above but negated using the ! ("not") modifier.
• A set of addresses using a list.
• The keyword any meaning all addresses

src_port

The source port in the Layer 4 packet header. Ports can be specified as:

• A number between 1 and 65535
• A valid service name from /etc/services
• A set of ports using a list
• A range:
  • != (not equal)
  • < (less than)
  • > (greater than)
  • <= (less than or equal)
  • >= (greater than or equal)
  • >< (range)
  • <> (inverse range)

  The last two are binary operators (they take two arguments) and do not include the arguments in the range.

The port option is not usually used in nat rules because the goal is usually to NAT all traffic regardless of the port(s) being used.

dst_addr

The destination address of packets to be translated. The destination address is specified in the same way as the source address.

dst_port

The destination port in the Layer 4 packet header. This port is specified in the same way as the source port.

ext_addr

The external (translation) address on the NAT gateway that packets will be translated to. The external address can be specified as:

• A single IPv4 or IPv6 address.
• A CIDR network block.
• A fully qualified domain name that will be resolved via DNS when the ruleset is loaded.
• The name of the external network interface. Any IP addresses assigned to the interface will be substituted into the rule at load time.
• The name of the external network interface in parentheses ( ). This tells PF to update the rule if the IP address(es) on the named interface changes. This is highly useful when the external interface gets its IP address via DHCP or dial-up as the ruleset doesn't have to be reloaded each time the address changes.
• The name of a network interface followed by the :network keyword. The resulting CIDR network (e.g. 192.168.0.0/24) will be substituted into the rule when the ruleset is loaded.
• A set of addresses using a list.

This would lead to a most basic form of this line similar to this:

nat on tl0 from 192.168.1.0/24 to any -> 24.5.0.5

This rule says to perform NAT on the tl0 interface for any packets coming from 192.168.1.0/24 and to replace the source IP address with 24.5.0.5.

While the above rule is correct, it is not recommended form. Maintenance could be difficult as any change of the external or internal network numbers would require the line be changed. Compare instead with this easier to maintain line (tl0 is external, dc0 internal):

nat on tl0 from dc0/24 to any -> tl0

The advantage should be fairly clear: you can change the IP addresses of either interface without changing this rule.

When specifying an interface name for the translation address as above, the IP address is determined at pf.conf load time, not on the fly. If you are using DHCP to configure your external interface, this can be a problem. If your assigned IP address changes then NAT will continue translating outgoing packets using the old IP address. This will cause outgoing connections to stop functioning. To get around this, you can tell PF to automatically update the translation address by putting parentheses around the interface name:

nat on tl0 from dc0/24 to any -> (tl0)

There is one major limitation to doing this: Only the first IP alias on an interface is evaluated when the interface name is placed in parentheses.

Bidirectional Mapping (1:1 mapping)

Example:

web_serv_int = "192.168.1.100"
web_serv_ext = "24.5.0.6"

binat on tl0 from $web_serv_int to any -> $web_serv_ext

Translation Rule Exceptions

no nat on tl0 from 192.168.1.10 to any
nat on tl0 from 192.168.1.0/24 to any -> 24.2.74.79

Then the entire 192.168.1.0/24 network would have its packets translated to the external address 24.2.74.79 except for 192.168.1.10.

Note that the first matching rule wins; if it's a no rule, then the packet is not translated. The no keyword can also be used with binat and rdr rules.

Checking NAT Status

   # pfctl -s state
   TCP 192.168.1.35:2132 -> 24.5.0.5:53136 -> 65.42.33.245:22 TIME_WAIT:TIME_WAIT
   UDP 192.168.1.35:2491 -> 24.5.0.5:60527 -> 24.2.68.33:53   MULTIPLE:SINGLE

Explanations (first line only):

TCP

The protocol being used by the connection.

192.168.1.35:2132

The IP address (192.168.1.35) of the machine on the internal network. The source port (2132) is shown after the address. This is also the address that is replaced in the IP header.

24.5.0.5:53136

The IP address (24.5.0.5) and port (53136) on the gateway that packets are being translated to.

65.42.33.245:22

The IP address (65.42.33.245) and the port (22) that the internal machine is connecting to.

TIME_WAIT:TIME_WAIT

This indicates what state PF believes the TCP connection to be in.

[Previous: Queueing] [Contents] [Next: Traffic Redirection]