CIDR - networking

[James Knott]

Kihara Muriithi wrote:
>> Then again, given the scarcity of non-CIDR systems these days, why are
>> you so concerned with it?  In the 12 years I've been working with IP,
>> I've never seen anything but CIDR.  Address classes have been obsolete
>> for years.
>   That sentence has homed at the problem. Its that I can't tell how a
> class address look like. My understanding was that, if you see a
> netmask like 255.0.0.0 or 255.255.0.0 on what they were calling A and
> B respectively, then that is class system. That is how my box is
> currently set up. ie 192.168.1.3  255.255.255.0
>   I thought CIRD carry a mask like 255.128.0.0? ie no clean 255 to 0
> transition. How would I identify a non-CIDR if I see one? I am very
> sorry for my ignorance by the way
> 
> William
> 
> On 1/8/06, James Knott <james.knott-bJEeYj9oJeDQT0dZR+AlfA at public.gmane.org> wrote:
>> Kihara Muriithi wrote:
>>> Hi
>>>  Thanks for the response. I suspected most modern host will work with
>>> CIRD, but haven't seen it used widely. Since it has been over 10 years
>>> since CIRD came to existance, this should not be the case and I
>>> wondered whether there was a technical reason behind it and hence the
>>> email
>>>  Now Knott, since you have experience with this set up, have you came
>>> across a problem that you can attribute to CIRD misconfiguration? What
>>> were the symptoms, just in case I ever happen to come across that
>>> issue? Allen mentioned an instance where CIRD can not be used and this
>>> may lead to a situation where classed and classless system network.
>>>  One last thing, am I correct to assume one identify whether a host is
>>> configured with class or classless system by using "ifconfig" and
>>> "route" commands and then looking at the netmask?
>> As I understand it, if you have a subnet mask where you can specify how
>> many bit to use, you have CIRD.  As I don't ever recall working with a
>> non CIDR system, I can't speak to any problems caused by it.  Bear in
>> mind, that any CIDR system can be configured to behave like a non CIDR
>> system.

The fact that a netmask falls on one of the class boundaries does not
mean it's using address classes.  What makes it CIDR, is the ability to
use *ANY* practical netmask.  Some of the valid masks will, of course,
appear to be one of the old classes.
--
The Toronto Linux Users Group.      Meetings: http://tlug.ss.org
TLUG requests: Linux topics, No HTML, wrap text below 80 columns
How to UNSUBSCRIBE: http://tlug.ss.org/subscribe.shtml