Class C subnetting and Novell TCP/IP

In <9403230932.AA00251 at buche>, <bob at informatics.rutherford.ac.uk> wrote:

> To all those that replied: many thanks, we seem to have unanimous agreement
> that subnetting on any (legal) mask is indeed possible, and that perhaps
> it's more a documentation/education problem.
> 
> Keith's taking this back to those customers of his who started this
> particular hare running.

Well, I just have, and they were indeed trying to break the
subnetting rules for Class Cs, this was their problem. Suddenly
their existing allocation became enough.

Thanks to everyone for clearing this up, particulary the JNT for usefully
re-circulating my request, and Andrew Gregg at University of Ulster
for hard exerience.

In <9403221550.AA09947 at reif.ripe.net>, <Daniel.Karrenberg at ripe.net> wrote:

>   > bob at informatics.rutherford.ac.uk writes:

>   > (b) if said customers persist perhaps Daniel could pursue this with his
>   >     Novell contact to get a definitive statement?
> 
> The customer's answer notwithstanding I have already started on b).

My apologies for casting aspersions at Novell - if Daniel does have a
word with them, it should probably be to the effect their
documentation needs to be idiot-proofed a bit better.

Keith

(I have a short one-page guide which describes all the legal
 subnetting options for Class Cs that we give to customers. If there
 was interest, it might be a good idea to expand this to include the
 valid host address ranges associated with each mask, and make it
 available.)