SUMMARY : Modem callback tools

From: Steffen Grunewald (steffen@gfz-potsdam.de)
Date: Thu Oct 21 1993 - 04:45:20 CDT


I wrote :
|> From steffen Mon Oct 4 15:54:39 1993
|> Subject: Q: Modem callback tools ?
|> To: sun-managers@eecs.nwu.edu
|> Date: Mon, 4 Oct 1993 15:54:39 +0100 (MET)
|> X-Mailer: ELM [version 2.4 PL21]
|> Content-Type: text
|> Content-Length: 743
|>
|> Hello helping hands out there,
|>
|> I'm looking for a piece of code (shell scripts and/or C) to
|>
|> - let users log in via a modem line and identify
|> - log them out automatically
|> - call a given number (which is associated to the user's identification)
|>
|> The modem will be connected to /dev/ttya on a Sun Sparc running SunOS 4.1.x.
|>
|> I didn't have to manage modems before, so be kind to a newbie (at least
|> in that field :). (Something I might have overlooked in TFM ?)
|>
|> Any hints are welcome. Summary next weekend. Thank you.
|> --
|> Steffen Grunewald | email steffen@gfz-potsdam.de | phone (+49)-331-310526
|> GeoForschungsZentrum Potsdam, Telegrafenberg A3, D-14773 Potsdam, Germany
|> #include <disclaimer.h> /* .sig: Permission denied */
|>

Sorry for the delay, but I wanted to give it a try before summarizing...

Some people said I should get one of these nice modems which do
calling-back in hardware. Because _I_already_have_a_modem_ (and
not such a beast), that was of no help.

I got a hint to use modempool by Lars Berntzon, too. The uuencoded
file that was included into that mail was not the actual version,
but I had a look at it and found that I will need at least two modems
to have a free line to call back (did I understand it right ?).
Large package, many whistles and bells for my purpose, I think...

Steve Beaty offered a short C program acting as getty login replacement.
I was tweaking this heavily, but had no success until now.

So the best solution would be : drop the available modem and get
an intelligent one (in fact, I could use the `old' thingy at home :/ )
or get into the logic of init/getty/login and write something yourself :(

Another way, avoiding getty, would be polling the serial port in question,
waiting for an incoming call, grabbing the line, asking for identification
(very similar to Steve's solution, BTW) and then hanging up, to call
again and spawn a getty to the already connected line...

Any hints ?

Thank you, whoever responded (or will respond).

I counted the following contributors :

From: fmufti@genetics.bio-rad.com (Fazeel Mufti 6895)
From: hawk!carl@netcom.com (Carl Gabrielson )
From: crm@lgi.com (Charlie Mengel - LGI)
From: mike_p <mike_+a_WTL_+lmike+_p+r%Trans_Ocean_Express@mcimail.com>
From: burdick@eagle.mcclellan.af.mil (Jim Burdick)
From: etnsed!xhaque@uunet.UU.NET (Amanul Haque)
From: Peter.Samuel@nms.otc.com.au (Peter Samuel)
From: beaty@paradox.lance.colostate.edu (steve beaty)
From: Pascal Meheut <pascal@cnam.cnam.fr>

My apologies, if I have missed someone's answer...

See you later, Steffen

-- 
Steffen Grunewald | email steffen@gfz-potsdam.de | phone (+49)-331-310526
GeoForschungsZentrum Potsdam, Telegrafenberg A3, D-14773 Potsdam, Germany
#include <disclaimer.h>        /*       .sig: Permission denied        */



This archive was generated by hypermail 2.1.2 : Fri Sep 28 2001 - 23:08:26 CDT