From owner-freebsd-questions  Wed Oct 16 20:14:53 1996
Return-Path: owner-questions
Received: (from root@localhost)
          by freefall.freebsd.org (8.7.5/8.7.3) id UAA28006
          for questions-outgoing; Wed, 16 Oct 1996 20:14:53 -0700 (PDT)
Received: from lserver.infoworld.com (lserver.infoworld.com [192.216.48.4])
          by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id UAA27999
          for <questions@freebsd.org>; Wed, 16 Oct 1996 20:14:50 -0700 (PDT)
From: BRETT_GLASS@infoworld.com
Received: from ccgate.infoworld.com (ccgate.infoworld.com [192.216.49.101]) 
	by lserver.infoworld.com (8.7.5/8.7.3/GNAC-GW-1.2) with SMTP id UAA03244 
	for <questions@freebsd.org>; Wed, 16 Oct 1996 20:14:43 -0700 (PDT)
Received: from ccMail by ccgate.infoworld.com (SMTPLINK V2.11)
	id AA845521883; Wed, 16 Oct 96 20:20:40 PST
Date: Wed, 16 Oct 96 20:20:40 PST
Message-Id: <9609168455.AA845521883@ccgate.infoworld.com>
To: questions@freebsd.org
Subject: Programming question: How to identify owner of a socket?
Sender: owner-questions@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

I'm writing a TCP/IP daemon (initially in Perl; probably in C
eventually) to run on a FreeBSD machine which functions as a terminal
server.  The users who dial in on the modems use SLiRP to connect via
PPP or SLIP. (This means that they will all appear to have the same IP
address: the IP address of the local host.)

For security reasons, I'd like the daemon to accept requests from
users on the modems only -- not from across the Net. I'd also like it to be
able to identify the user making each request.

To do this, the daemon has to verify that the client's IP address
matches that of the local host. It should then verify that the client's IP
port number belongs to a local instance of SLiRP.  Finally, it should
identify the user by looking up the owner of the SLiRP process.

I'm sure all of this is possible. However, since I'm not experienced at
building IP daemons under UNIX, I don't know quite how to do it. What are
the system calls that:

Let a daemon find the IP address of the connecting client?

Let it find the IP port number of the connecting client?

Map the IP port number to a process and user on the local machine?

And can these all be called from Perl, so I can do the first draft of the
daemon without coding it in C?

Any help will be much appreciated.

--Brett Glass