Discussion:
PPP Protocol Development From Scratch
(too old to reply)
Mahesh
2008-07-10 04:59:35 UTC
Permalink
Hi All,

I wanted to develop PPP Protocol From Scratch just for hobby. Dont ask
me "why reinvent wheel" .

I am crazy for netowkring so wanted to explore more on protocol
development.

I request any of one here to help regarding how to go about or just
give me pdf/doc that explains a bit for development.

Waiting for +ve response

Thanks a lot.

-
Mahesh
R.L. Horn
2008-07-10 05:57:48 UTC
Permalink
Post by Mahesh
I wanted to develop PPP Protocol From Scratch just for hobby. Dont ask
me "why reinvent wheel" .
I am crazy for netowkring so wanted to explore more on protocol
development.
I request any of one here to help regarding how to go about or just
give me pdf/doc that explains a bit for development.
The protocol and extensions are described in numerous IETF RFCs
(http://www.ietf.org/rfc.html). I'd start with RFC 1661 and go from there.
Most of the other relevant documents can be found through the RFC index or
through the usual search engines (search for "RFC" plus whatever you're
interested in, e.g. PAP, CHAP, etc.) and, believe it or not, Wikipedia.

For nitty-gritty implementation details, there's always the pppd source
(http://www.samba.org/ppp/).
Unruh
2008-07-10 16:24:45 UTC
Permalink
Post by Mahesh
Hi All,
I wanted to develop PPP Protocol From Scratch just for hobby. Dont ask
me "why reinvent wheel" .
No idea what you are talking about. If you redevelop the protocol, you do
not have ppp. You have something else. If you mean you want to impliment
ppp from the protocol, then the protocol is defined in the appropriate RFC
(I do not have the number unfortunately). However, that is actualy
pointless because there are loads of ppp implimentations out there which
actually break the protocol, and you also need to make your ppp
implimentation robust enough that you can handle those non-compliant
versions as well. That is why it is silly to try to "roll your own" for
anything useful. For learning it may be useful. There is also a book by
Carlson which descibes ppp is gory detail.
Post by Mahesh
I am crazy for netowkring so wanted to explore more on protocol
development.
I request any of one here to help regarding how to go about or just
give me pdf/doc that explains a bit for development.
Waiting for +ve response
Thanks a lot.
-
Mahesh
R.L. Horn
2008-07-10 19:26:04 UTC
Permalink
...However, that is actualy pointless because there are loads of ppp
implimentations out there which actually break the protocol, and you also
need to make your ppp implimentation robust enough that you can handle
those non-compliant versions as well. That is why it is silly to try to
"roll your own" for anything useful.
Actually, I can think of a number of reasons why it wouldn't be "silly," but
I consider "because it sounds like fun" to be perfectly adequate. Why
discourage the guy, especially since the original post suggests that he
already knows the project may be a bit quixotic?
Mahesh
2008-07-11 02:41:38 UTC
Permalink
Post by R.L. Horn
...However, that is actualy pointless because there are loads of ppp
implimentations out there which actually break the protocol, and you also
need to make your ppp implimentation robust enough that you can handle
those non-compliant versions as well. That is why it is silly to try to
"roll your own" for anything useful.
Actually, I can think of a number of reasons why it wouldn't be "silly," but
I consider "because it sounds like fun" to be perfectly adequate.  Why
discourage the guy, especially since the original post suggests that he
already knows the project may be a bit quixotic?
R.L. Horn: Thank you so much for your great assist. :)
Unruh
2008-07-11 15:19:51 UTC
Permalink
Post by R.L. Horn
...However, that is actualy pointless because there are loads of ppp
implimentations out there which actually break the protocol, and you also
need to make your ppp implimentation robust enough that you can handle
those non-compliant versions as well. That is why it is silly to try to
"roll your own" for anything useful.
Actually, I can think of a number of reasons why it wouldn't be "silly," but
I consider "because it sounds like fun" to be perfectly adequate. Why
discourage the guy, especially since the original post suggests that he
already knows the project may be a bit quixotic?
Because he may then be tempted to make it not just for fun, but foist it
upon an unsuspecting public. That is why the "anything useful" caveate is
in there. I also tried to supply him with some hints as to where to look.
Loading...