From owner-freebsd-questions  Sun Dec 22 15:10:22 1996
Return-Path: <owner-questions>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.4/8.8.4) id PAA24856
          for questions-outgoing; Sun, 22 Dec 1996 15:10:22 -0800 (PST)
Received: from seagull.rtd.com (seagull.rtd.com [198.102.68.2])
          by freefall.freebsd.org (8.8.4/8.8.4) with ESMTP id PAA24850
          for <freebsd-questions@freebsd.org>; Sun, 22 Dec 1996 15:10:20 -0800 (PST)
Received: (from dgy@localhost) by seagull.rtd.com (8.7.5/8.7.3) id QAA08932; Sun, 22 Dec 1996 16:10:13 -0700 (MST)
From: Don Yuniskis <dgy@rtd.com>
Message-Id: <199612222310.QAA08932@seagull.rtd.com>
Subject: Re: your mail
To: drussell@internode.net (Doug Russell)
Date: Sun, 22 Dec 1996 16:10:13 -0700 (MST)
Cc: freebsd-questions@freebsd.org
In-Reply-To: <1.5.4.32.19961222212543.007051e4@internode.net> from "Doug Russell" at Dec 22, 96 02:25:43 pm
X-Mailer: ELM [version 2.4 PL24]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Sender: owner-questions@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

> >That's *NOT* the correct thing.  What happens when for some reason
> >PPP happens to send the sequence '+++' to the modem?  All of a sudden
> >it'll drop into command mode and you're screwed.  User-PPP (as well as
> >all other PPP/SLIP implementations I've worked with) assumes that you've
> >disabled the escape sequence at least temporarily.
> 
> You aren't ENTIRELY correct.  If set up properly, the +++ sequence SHOULD
> work properly.  a +++ in the datastream won't trigger going into command
> mode unless there has been no data sent for a specified delay period.  (See
> S12 in your modem's S-Rigister table.  From a Courier HST V/Everthing:  Sets
> the duration, in fiftieths of a second, of the guard time for the escape
> code (+++) sequence.)  Default is 50 (1 second)).  So, in other words, you
> have to not just have a +++ come over the connection, but at least a one
> second pause,  a +++ (and only a +++) then another 1 second or longer pause,
> in order for the modem to go into command mode.  It is possible, but highly
> unlikely.

Easier to just disable the +++ sequence -- change your "escape character".
I don't like things that are "highly unlikely" -- that tends to mean
their probability of occuring is damn close to 100%!  :-(

> Of course, this is not the "proper" way of hanging up a modem.  The
> generally accepted method (especially with UNIX systems) is to drop DTR.
> The modem should be set to at least hang up, if not do a full reset, on DTR
> drop.
> 
> I have always set my modems this way under DOS, and under UNIX.
> 
> Biggest reason with a good old term program in DOS?  It takes MUCH less time
> to do a vDTR^ than it does to wait a second, +++, wait a second, ATH.

Yes.  ALso there are other things besides PC's (etc.) in the world
that actually *use* these signaling conventions!

--don