Matching Address Classes with Prefix-Lists (Followup)

This article is a follow up on the Matching Address Classes with Prefix-Lists based on the comments from killerkadoogan. I think its easier to answer the comments this way, I hope you don’t mind 🙂

Hi Pashtuk,

Under the technical definition of a CLASS C network, IE using classes. is NOT a class C network. So ‘Class C: le 32′ would also block or allow that ‘supernet’ of Class C’s, which isn’t a Class C in itself. Old-school routers (ie RIPv1) would convert the mask to /24,Class C is technically:
bits 1-3 110 + a 24 bit mask.


check out“in class c, the high order three bits are one-one-zero, the next 21 bits are the network and the last 8 bits are the local address.”so the technically correct Class C prefix list is: ge 24 le 24

you wouldn’t use that in the real world though ;)

You are right if we are talking about the old “classical” concept of Address Classes with their definition of the subnet mask.

Nonetheless the classical concept does not really matter much anymore nowadays, even though the terms Class A/B/C are still in use. Most of the time if we talk about a Class C network today a /24 subnet is thought off, no matter to which original Class the subnet belongs too.

On another occasion we can also talk about Classes in the meaning of ranges grouped into those defined classes (as used in the article).  In that case, it doesn’t matter what the typical subnet mask was, it only matters that we are able to match every subnet out of that range no matter which mask it got.

It does not have much to do with a real world approach, as a lot of the CCIE stuff I had to learn, but it belongs into the stuff a person has to know for the CCIE lab.



  1. killerkadoogan

    Hi Pashtuk,

    You should drop the use of the word ‘Class’ then, because the RFC definition is correct.
    You’re using ‘classical’ terminology to describe something that does not fit the ‘classical’ definition – and applying only half of the criteria, ie that the first 3 bits are 110, but not the second – that the mask should be 24 bits.

    Add auto-summary to a routing protocol and we see it apply the 24 bit class-full mask, for instance:

    interface Loopback3
    ip address

    router bgp 1

    #sh ip bgp
    BGP routing table entry for, version 18

    You are correct when you say your prefix will catch all of the Class C addresses, but it will also catch addresses that a *router* would not consider Class C.

    Or Jeff Doyle for that matter (from TCP IP routing Volume 2)..

    “Summarization can also cross class boundaries. For example, the four Class C networks (,,, and can all be summarized wit the aggregate address Notice that the aggregate, with its 22-bit mask, is no longer a legal Class C address.”

    It is more important to understand how a *router* applies class – than whatever a popular current misconception of the term is.



  2. pashtuk

    Heya ,-)
    Do you have a better wording which fits the same as Address Classes and is clear for everyone what is being talked about?

    It is also not about how a router is going to apply stuff with active auto-summary, its just on how to match those IP Ranges, which are classified as Class A, B and C, in a classless manner. Classes with Classfull networks are more or less dead nowadays, but the terms are still used to explain which IP range we are talking about. Sorry ,-)

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s