[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: [H.323] H.323 status



At 08:39 PM 12/3/97 -0500, Adam Heath wrote:
>First, a little history.
>
>As I am sure we all know, Linux currently does not support masquerading of
video
>conferencing protocols, including H.323.  There are partial solutions using
>IPPORTFW, but none that work transparently.
>
>So, a few weeks ago I started research into this project.  I found that doing
>this was not going to be easy.  The initial connection is done on a
well-known
>port.  During that communication, a dynamic control channel is
netgotiated, and
>that channel is then used to allocate 'logical' channels, for video, data,
and
>audio, using UDP.
>
>Normally, this wouldn't be a BIG problem, but the first two control
channels are
>encoded with ASN.1/PER.  For those of you that don't know what that is, it
is a
>data definition language, that allows for transfer of data between disparate
>machines.  Also, it allows of OPTIONAL records in the data stream, which
makes
>it hard to just extract addresses at specific offsets.
>
>In my research, I had found the ASN.1 definitions for H.323, but didn't
have a
>compiler to convert them into C header files and translation functions.
When I
>went to find one, the only good one I found was SNACC, which used BER,
instead
>of PER.

ASN decoding, lots of fun. I once wrote one of these for an SNMP module for
an embedded print server many years back. Unfortunately, I don't have
access to that code anymore. I don't remember what the BER/PER stuff is, I
thought it was all pretty generic. I would think you could find source for
an ASN decoder in the Linux SNMP code. Might be what you are looking for.

[snip]

>First, I will be making a masquerading module, which will only support
outgoing
>connections.  Then, I will implement a user-mode proxy, that will comunicate
>with the kernel module, and allow incoming connections.  I also want to
make a
>device driver to allow access to the data, audio, and video in an easier way.
>

Sounds great!

>I plan on having a basic kernel implementation by the end of the year.
>
>Wish me luck+ACE-  :)
>

Good luck.

>
> Adam Heath of Borg-Linux adam.heath+AEA-usa.net Join the H323 effort.  Email
> http://www.debian.org - Get Your Own Linux+ACE-
h323-request+AEA-cichlid.com with
> http://wwp.mirabilis.com/3375265 - Page Me  the word subscribe in the body.
>
> Windows 95: --  32-bit extensions and a graphical   Windows v. Linux is
> shell for a 16-bit patch to an 8-bit operating      a no-win situation.
> originally coded for a 4-bit micropro-
> cessor written by a 2-bit company that   It is nearly impossible to
> can't stand for 1 bit of competition.    look at a penguin and get angry.
>
>ps.  The linux kernel comes with source(duh+ACE-), but I was wondering
what was
>thought about distributing the documentation that was used to generate the
free
>source.
>
>
>-
>To unsubscribe use h.323-request@cichlid.com
>Archives http://www.scruz.net/~cichlid/h.323-archive
>

Al Youngwerth
alberty@apexxtech.com


--
TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to
debian-devel-request@lists.debian.org . 
Trouble?  e-mail to templin@bucknell.edu .


Reply to: