[PLUG] CIDR in ipv4 addressing questions...

[Kris]

On Mon, Apr 14, 2003 at 10:05:48AM -0700, Michael Robinson wrote:
> There's an RFC , I believe 1519, that defines Private class C
> networks in CIDR notation.  How is 192.168.0.0/16 a supernet for class C
> networks?  I believe that the latter idea of classes was based on ranges
> of  octet bound numbers where the current idea is based on how many 
> bits are in the subnet mask.  This isn't completely correct through 
> because certain numbers are known to be special such 
> as 192.0.2.0/somenumber which is supposed to be used for test 
> nets and 0.0.0.0/0 which is a broadcast special address.  Help.

CIDR = Classless InterDomain Routing

The key word is "classless".  Classless routing allows us to do cool
things like supernetting, subnetting, and non-standard'netting. :)

The RFC defines the 192.168 space as a supernet: 192.168.x.x is really
what it is saying.  This means that 192.168.47.0/24 and 192.168.0.0/24
are both in a private range.  If you saw 192.168.0.0/16 in a routing
table, it is most likely a summary route (all 192.168.x.x routes.. go
this way).

> Will ipv6 have private addressing?  Is ipv6 addressing going to
> support ipv4 addressing as the lower half of an ipv6 number?
> How does ipv6 with 128 bit numbers not pose a problem for the
> routing infrastructure?  Will ipv6 be used at the home level or
> just on backbones, at research locations, and on certain
> campuses for training?  Is ipv6 restricted to hex notation
> for addresses?  Will ipv6 need CIDR or is it more flexible for
> getting the correct number of hosts for a particular network?

Can't help ya here.  I havent studied up on my IPv6.

-- 
I'm just a packet pusher.