Delegating class C's

[Joseph S D Yao]

On Wed, May 16, 2001 at 08:26:01AM -0700, Robert Gahl wrote:
...
> I've never had a good grasp on what constitutes a class C, class B or class 
> A address, just that people always gave me a "class C". I had thought that 
> the numbers used in the IP address dictated whether an address was class A, 
> B or C (based on a calculator I use to help me figure out netmasks, etc.). 
> I stand corrected and I apologize for my being obtuse.

No apologies needed.  These are somewhat outdated terms in this
context, and are now being used in a much looser fashion than their
"technically correct" definitions.  I hope you don't mind my cc'ing
this to the list, because you are probably not the only one who is just
trying to define them in context.

A "Class A" network is one of the networks whose first byte is in the
range 0-127.  It USED TO BE defined always to have an 8-bit netmask,
255.0.0.0, hence a lot of people now call any network with that netmask
a "Class A", even if it's 255.0.0.0/8.  

This is now referred to as a network with an 8-bit netmask.  ;-)

Since many of the networks whose first byte was above 127 are already
broken up, you will rarely see a /8 network that is not a real Class A
network, except (a) when you're dealing with reverse DNS lookups, and
(b) some of the reserved or formerly-reserved networks in the 128-255
range.

A "Class B" network is one of the networks whose first byte is in the
range 128-191.  It USED TO BE defined always to have a 16-bit netmask,
255.255.0.0, hence a lot of people now call any network with that
netmask a "Class B".  This is now referred to as a network with a
16-bit netmask.  ;-)

A "Class C" network is one of the networks whose first byte is in the
range 192-223.  It USED TO BE defined always to have a 24-bit netmask,
255.255.255.0, hence a lot of people now call any network with that
netmask a "Class C".  This is now referred to as a network with a
24-bit netmask.  ;-)

The rest are "Class D", used for multi-cast, and "Class E" addresses.
IIRC, Class D is 224-239, and Class E is 240-255.

See also <URL: http://www.iana.org/assignments/ipv4-address-space>.

To quote an old RFC:

-----------------------------------------------------------------------
   3. If the first bit of the first octet of the address is 0 (zero),
   then the next 7 bits of the first octet indicate the network number
   (Class A Address).  If the first two bits are 1,0 (one,zero), then
   the next 14 bits define the net number (Class B Address).  If the
   first 3 bits are 1,1,0 (one,one,zero), then the next 21 bits define
   the net number (Class C Address) [See RFC-943].
-----------------------------------------------------------------------

The current descendant of RFC 943 is RFC 1700, which says:

-----------------------------------------------------------------------
Special Addresses

There are five classes of IP addresses: Class A through Class E.  Of
these, Classes A, B, and C are used for unicast addresses, Class D is
used for multicast addresses, and Class E addresses are reserved for
future use.

With the advent of classless addressing [CIDR1, CIDR2], the
network-number part of an address may be of any length, and the whole
notion of address classes becomes less important.
-----------------------------------------------------------------------

I hope this helps!  ;-)

-- 
Joe Yao				jsdy at cospo.osis.gov - Joseph S. D. Yao
OSIS Center Computer Support					EMT-B
-----------------------------------------------------------------------
This message is not an official statement of COSPO policies.