Configuration Guide Configuring ARP
2 Configuring ARP
2.1 Overview
In a local area network (LAN), each IP network device has two addresses: 1) local address. Since the local address is
contained in the header of the data link layer (DLL) frame, it is a DLL address. However, it is processed by the MAC
sublayer at the DLL and thereby is usually called the MAC address. MAC addresses represent IP network devices on
LANs. 2) network address. Network addresses on the Internet represent IP network devices and also indicate the
networks where the devices reside.
In a LAN, two IP devices can communicate with each other only after they learn the 48-bit MAC address of each other.
The process of obtaining the MAC address based on the IP address is called address resolution. There are two types of
address resolution protocols: 1) Address Resolution Protocol (ARP); 2) Proxy ARP. ARP and Proxy ARP are described
respectively in RFC 826 and RFC 1027.
ARP is used to bind the MAC address with the IP address. When you enter an IP address, you can learn the
corresponding MAC address through ARP. Once the MAC address is obtained, the IP-MAC mapping will be saved to the
ARP cache of the network device. With the MAC address, the IP device can encapsulate DLL frames and send them to
the LAN. By default, IP and ARP packets on the Ethernet are encapsulated in Ethernet II frames.
Protocols and
Standards
RFC 826: An Ethernet Address Resolution Protocol
RFC 1027: Using ARP to implement transparent subnet gateways
2.2 Applications
A user learns the MAC addresses of other users in the same network segment
through ARP.
Proxy ARP-based Transparent
Transmission
With Proxy ARP, a user can directly communicate with users in another network
without knowing that it exists.
2.2.1 LAN-based ARP
Scenario
ARP is required in all IPv4 LANs.
A user needs to learn the MAC addresses of other users through ARP to communicate with them.