I have an awesome book for you! You must read Radia Perlman's
Interconnections. This book not only digs into this issue, but relates IP to
IPX, CNLP, DEC, and IPv6. I will first add we would not be able to call CIDR
address listings in routing tables prefixes if non-contiguous masks were
ever allowed in any vendors software. I was approached with this very
question a little while ago I had just completed my Microsoft TCP/IP test
and I thought I knew subnetting. A friend of mine asked why the masks had to
be contiguous and I had a really dumb answer for him which was- that is just
how it is. I will try to sum up what Radia goes over why this is so. If you
think about this first and foremost non-contiguous subnet masks would make a
confusing system even more confusing reading. In the beginning there wasn't
anything keeping a vendor from implementing software that would allow
non-contiguous subnet masks, but amazingly enough nobody was stupid enough.
This would have made a not so efficient protocol even more inefficient. This
would absolutely require the router to read the entire address before any
routing decisions were made~ ouch. Then what is really neat is how Radia has
homework at the end of the chapter to help bring to light how difficult it
would be to determine if 2 addresses are in the same subnet. (the book is
awesome). Luckily non-contiguous masks were never implemented and we can
call them darn slash things (/28) prefixes, our routing tables are smaller,
and our processors have less load (humans and routers).
>>>Brian
>From: "Tony van Ree" <[EMAIL PROTECTED]>
>Reply-To: "Tony van Ree" <[EMAIL PROTECTED]>
>To: "J. Kata" <[EMAIL PROTECTED]>, [EMAIL PROTECTED]
>Subject: Re: Non-contiguous subnet masks
>Date: Mon, 22 Jan 2001 10:07:45 +1100
>
>Hi,
>
>I'm old and sometimes get confused. Why would one want to do something
>that is not really very standard even if they could. I once used to do a
>lot of troubleshooting for a group that never had time to standardise their
>network because it was "unique" and therefore required all their time to
>manage it. It was "unique" because it was not standardised.
>
>Don't most big failures occur because we do something others wouldn't?
>
>If so, why would you use no contiguous subnetmasks rather than blockout
>nicely what you need? Is there a substitute for good design?
>
>Teunis
>Hobart, Tasmania
>Austalia
>
>
>On Saturday, January 20, 2001 at 07:19:19 PM, J. Kata wrote:
>
> > Supposedly, non-contiguous subnet masks are
> > legal because the rfc's don't restrict their
> > use. Are there any good resources where I can
> > read up on this subject? Has someone come across
> > their use in the real world?
> >
> > And finally, can anyone answer these questions?
> >
> > IP: 172.31.100.100
> > SM: 255.255.24.164
> >
> > What subnet does this ip address belong to?
> > How many subnets does this mask support?
> > How many hosts does this mask support?
> > What is the broadcast address?
> >
> > Thanks.
> >
> > -- Janusz Kata
> >
> > _________________________________
> > FAQ, list archives, and subscription info:
>http://www.groupstudy.com/list/cisco.html
> > Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
> >
> >
>
>
>--
>www.tasmail.com
>
>
>_________________________________
>FAQ, list archives, and subscription info:
>http://www.groupstudy.com/list/cisco.html
>Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
_________________________________________________________________
Get your FREE download of MSN Explorer at http://explorer.msn.com
_________________________________
FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
