From owner-freebsd-current  Wed Sep 10 22:28:22 1997
Return-Path: <owner-freebsd-current>
Received: (from root@localhost)
          by hub.freebsd.org (8.8.7/8.8.7) id WAA27969
          for current-outgoing; Wed, 10 Sep 1997 22:28:22 -0700 (PDT)
Received: from word.smith.net.au (word.smith.net.au [202.0.75.3])
          by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id WAA27960
          for <current@freebsd.org>; Wed, 10 Sep 1997 22:28:19 -0700 (PDT)
Received: from word.smith.net.au (localhost.smith.net.au [127.0.0.1])
	by word.smith.net.au (8.8.7/8.8.5) with ESMTP id OAA06582;
	Thu, 11 Sep 1997 14:56:13 +1000 (EST)
Message-Id: <199709110456.OAA06582@word.smith.net.au>
X-Mailer: exmh version 2.0zeta 7/24/97
To: Simon Shapiro <Shimon@i-Connect.Net>
cc: Terry Lambert <tlambert@primenet.com>, current@freebsd.org,
        nnd@itfs.nsk.su
Subject: Re: PPP - why set CLOCAL for server ('-direct' mode) ? 
In-reply-to: Your message of "Wed, 10 Sep 1997 21:35:23 MST."
             <XFMail.970910213523.Shimon@i-Connect.Net> 
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Date: Thu, 11 Sep 1997 14:56:13 +1000
From: Mike Smith <mike@smith.net.au>
Sender: owner-freebsd-current@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

> 
> The solution is very simple;  Upon open, you open anyway, ignoring but
> remembering DCD.  If you get a change of state from inactive to active, you
> change the state, but ignore it.  However, if the state  transitions from 
> ACTIVE to INACTIVE, you terminate the connection, close the port and
> generate SIGHUP.  This is the simplest and most correct way to handle it.

This is how a callout device works, although SIGHUP is only delivered 
if the port is the controlling tty of the process.  Note, however, that 
user-mode ppp _doesn't_want_this_, which is the whole issue.  

It wants to poll for carrier status, hence CLOCAL.

mike