In "Hanging Out with the Classless Crowd" (January 2001), William Heldman states, "If you have subnets, you have a classless domain with a classless subnet mask." This statement isn't entirely accurate; you can have a subnetted network and not have a classless subnet mask.
In addition, the author gives the impression that you can simply subnet a network with a classless subnet mask and everything will work. This assumption isn't so.


What the author didn't mention in the article is the linchpin of classless routing--the routing protocols that you use.


If a company chooses to use Routing Information Protocol version 1 (RIPv1) or Interior Gateway Routing Protocol (IGRP) as its routing protocols, the subnet mask for the entire network would have to be uniform. These routing protocols don't include the subnet mask information in their routing updates. Because a
variable-length subnet mask (VLSM)
lets you subnet a subnet, this informa-tion is necessary. Routing protocols like Enhanced IGRP (EIGRP), Open Shortest Path First (OSPF), and border gateway protocol 4 (BGP-4) let classless routing take place.
*Jim Loprest

Jim Loprest 3/15/2001 10:43:43 AM

The way the sentence is worded sounds as though just having subnets automatically puts you into the classless domain or classless subnet mask genre. Conveying that information wasn't my intent, nor was presenting a primer about classless routing and routing protocols. All I was really going for in the article was to provide information at a fundamental level about how long subnet masks work and what slash notation is all about.

--Bill Heldman

Bill Heldman 3/15/2001 10:43:43 AM