User`s guide

36
On models with multiple Ethernet interfaces, this parameter may be used to specify a
maximum data rate in kbps that the unit will transmit on this interface. This may be useful
in applications where separate Ethernet interfaces are allocated to separate LANs and it is
necessary to prioritize traffic from one LAN over another.
TCP transmit buffer size
When set to a non-zero value, this parameter sets the TCP buffer size of transmitted
packets in bytes. This is useful for slow / lossy connections such as satellite. Setting this
buffer to a low value will prevent the amount of unacknowledged data from getting too high.
If retransmits are required, a smaller TX buffer helps prevent retransmits flooding the
connection.
Take this interface out of service after n seconds when the link is lost
(e.g. cable removed or broken)
This parameter is used to specify the length of time (in seconds) that the router will wait
after detecting that an Ethernet cable has been removed before routes that were using that
interface are marked as out of service. If the parameter is set to 0, the feature is disabled
i.e. routes using the interface will not be marked as out of service if the cable is removed.
Enable NAT on this interface
This parameter is used to select whether IP Network Address Translation (NAT) or Network
Address and Port Translation (NAPT) are used at the Ethernet interface. When the
parameter is set to disabled, no NAT will take place. When this parameter is enabled, extra
options described below will be displayed.
NAT and NAPT can have many uses but they are generally used to allow a number of private
IP hosts (PCs for example) to connect to the Internet through a single shared public IP
address. This has two main advantages, it saves on IP address space (the ISP only need
assign you one IP address), and it isolates the private IP hosts from the Internet (effectively
providing a simple firewall because unsolicited traffic from the Internet cannot be routed
directly to the private IP hosts.
To use NAT or NAPT correctly in the example of connecting private hosts to the Internet,
NAT or NAPT should be enabled on the router’s WAN side interface and should be disabled
on the router’s LAN side interface.
IP address
When a private IP host sends a UDP or TCP packet to an Internet IP address, the router
will change the source address of the packet from the private host IP to the router’s
public IP address before forwarding the packet onto the Internet host. Additionally it will
create an entry in a “NAT table” containing the private IP source address, the private IP
port number, the public IP destination address and the destination port number.
Conversely, when the router receives a reply packet back from the public host, it checks
the source IP, source port number and destination port number in the NAT table to
determine which private host to forward the packet to. Before it forwards the packet back
to the private host, it changes the destination IP address of the packet from its public IP
address to the IP address of the private host.
IP address and Port
This mode behaves like NAT but in addition to changing the source IP of the packet from
the private host it can also change the source port number. This is required if more than
one private host attempts to connect using the same local port number to the same
Internet host on the same remote port number. If such a scenario were to occur with
NAT the router would be unable to determine which private host to route the returning
packets to and the connection would fail.