Until the Dynamic Host Configuration Protocol (DHCP) was developed, automatic assignment of IP addresses to hosts was virtually obsolete. The only method available was BOOTP (RFC 0951), which required knowledge of the MAC layer address and preconfiguration of the IP address for a particular MAC on a BOOTP server before the station could get an address. If a station had to be moved from one subnet to another, the BOOTP server had to be reconfigured.
DHCP (RFC 2131/1541/1531) solved these problems by creating address scopes in subnets to be assigned automatically to DHCP clients. A scope is a range of subnet addresses used to dynamically assign IP addresses. This provides the benefit of being able to move a laptop from one subnet to another and be assigned an IP address.
Additional benefits are that other information, called DHCP options, can be automatically loaded to the PC client. DHCP options include subnet mask, default gateway, WINS server, Domain Name System (DNS) server, TFTP server, and others. Microsoft provides a DHCP server application as part of Windows NT. Cisco also provides a DHCP server called Cisco DNS/DHCP Manager.
A network designer must keep in mind the locations of DHCP servers in the network. Subnets without a local DHCP server require that the router forward the DHCP discover request to the DHCP servers. The ip helper-address command is used to accomplish this task. Looking at Figure 3-8, the interfaces Ethernet 0 and Ethernet 1 are configured to forward DHCP discover requests to the DHCP server.
The packet contains the source subnet information so that the server knows from which scope to assign IP addresses. Ethernet 2 does not need the ip helper-address command because the server is on the same subnet.
Figure 3-8 DHCP Traffic Flow
Interface Ethernet 0
IP address 10.1.2.1 255.255.255.0
Interface Ethernet 1
IP address 10.1.3.1 255.255.255.0
IP helper address 10.1.1.100
DHCP Server 10.1.1.100
Scope 1: 10.1.2.100-10.1.2.150 Scope 2: 10.1.3.100-10.1.3.150
Was this article helpful?