1. All the addresses in the CIDR address must be contiguous. Use of the standard network prefix notation for addresses,however, also makes it tidy and efficient to carve up any kind of address, as needed.
2. When address aggregation occurs, CIDR address blocks work best when they come in sets that are greater than 1 and equal to some lower-order bit pattern that corresponds to all 1s - namely in groups of 3, 7, 15, 31, and so on. That's because this makes it possible to borrow the corresponding number of bits (two, three, four, five,and so on) from the network portion of the CIDR address block and use them to extend the host portion instead.
3. To use a CIDR address on any network, all routers in the routing domain must "understand" CIDR notation. This is typically not a problem for most routers that were built after September 1993, when RFCs 1517, 1518, and 1519 were approved, because most router vendors began to support CIDR addresses at that time.
Here's a brief rundown of how DHCP works, from a client perspective:
1. When TCP/IP is configured on the client computer, the Obtain an IP address automatically option is the only necessary set-up element. The DHCP service is automatic, which explains the terrific appeal that DHCP holds for network administrators and users alike.
2. The next time the workstation attempts to access the network (older versions of Windows must be rebooted first), it broadcasts a DHCP address request to the network because it has no IP address. It can make this broadcast because it is now configured as a DHCP client.
3. All DHCP servers present on the same broadcast domain receive the request and send back a message that indicates a willingness to grant an address lease, if an address is available.
4. The client accepts an address lease offer (usually the first one it receives) and sends a packet to the server that extended the offer.
5. In reply, the server proffers an IP address for a specific period of time (which is why it's called a lease) that the client uses thereafter.
6. When half the lease period expires, the client attempts to renew the lease. Usually, the DHCP server that granted the lease will renew it, but if it doesn't respond, the client tries to renew again at other times during the lease period. Only if the client is unable to renew its lease before expiration must that client repeat the DHCP request process, as described in Step 2.
Server addresses (and sometimes their associated services) are advertised using DNS, which resolves domain names into IP addresses, and vice versa.
DNS is not a dynamic environment, so all address updates must be entered manually (either through a GUI interface, on Windows Server 2008, or by editing text files on UNIX systems).
Client addresses usually come into play only when e-mail addresses of the form email@example.com must be resolved. E-mail servers can resolve this information from MX records associated with the client's domain name (not his or her IP address), so dynamic address resolution works perfectly well for clients and e-mail. Hence, client addresses typically have no impact on DNS, or vice versa, and can change, as needed.