On Tue, 18 Jan 2000, der Mouse wrote:

> > People using noncontiguous netmasks are going to be used to a lot
> > worse than just long-line output. :-)
> 
> Like what?  I don't recall seeing anything break yet.  Everyone yammers
> about how noncontiguous netmasks breaks things, but my experience
> squarely contradicts that.

Some read the RFCs to imply it ins't allowed.. (I don't know if there is
basis for that or not)

Most routers are not tested with noncontiguous netmasks.  If it works you
are lucky, though expirence tends to suggest that it works anyway.

I have seen programs refuse to allow them.   MacTCP is a clasic example,
it did not allow you to set a noncontiguous netmask.  (it also did some
other very brain dead things in the name of helping out.  OpenTransport
is much nicer, fortunatly)  Gui interfaces to routers have been known not
allow the noncontiguous netmask.  Going in to the command line you could
manualy enter the "invalid" netmask, and it would take.  

Again, all this is not to say that you can't do it.  It is however rare,
untested, and confusing to most people who don't realise that it could be
done.  I cannot think of a good reason to use noncontiguous netmasks, I
sould discourage it unless you accually can come up with one.


-- 
      http://www.black-hole.com/users/henrymiller/ 
      hank@black-hole.com