Date: Mon, 11 Nov 1996 16:15:39 -0700 (MST) From: Terry Lambert <terry@lambert.org> To: neal@pernet.net (Neal Rigney) Cc: jgreco@brasil.moneng.mei.com, john@starfire.mn.org, hackers@freebsd.org, freebsd-isp@freebsd.org Subject: Re: PPP/LCP sensing getty Message-ID: <199611112315.QAA18862@phaeton.artisoft.com> In-Reply-To: <Pine.NEB.3.93.961111151541.4247A-100000@jennifer.pernet.net> from "Neal Rigney" at Nov 11, 96 03:17:35 pm
next in thread | previous in thread | raw e-mail | index | archive | help
> My modems > stupidly send a connect message in a way that getty sees as a username. > So it sits at Password: waiting for test, but only getting LCP. The other > times I got it to work(with different modems), it worked like a charm. Are you sure they are sending the message after raising DCD? It's been standard for years to send the connect messages *before* raising DCD. What might be happening is one of: o DCD is not set to follow remote carrier (modem problem) o DCD is not really DCD (cable problem) o You are using non-modem-control port for mode (/dev problem) o You are using a serial card that does not flush input buffer contents on off-to-on DCD transition on modem control ports (device driver problem) o Your modem is truly, honestly *bogus* (modem problem) I've never seen a truly bogus modem that supported the AT command set (the AT&T 4024 and the Tandy DC-15 are non-AT command set bogus modems; and the AT&T cn be firmware upgraded). You might also consider that, unless you are running the bogus uugetty (which should open with a flag to hang the open pending a "ring indicate" signal, and timeout on no DCD raid within XXX seconds of a ring indicate), response codes from the modem should be turned off anyway. They are only really useful for dialout progress reporting (before you tell me that you need them to recognize baud on a uugetty, let me point you to the RS232C and Bell 103C standards on external clocks). In any case, a uugetty or mgetty or whatever should flush input buffers before putting up a login, and should delay 1 second to ensure input buffer flush occurs, in any case, before outputting a prompt, in case the modem ignores RTS from the computer prior to raising its own DCD --- the "bogus modem" case, above). Regards, Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199611112315.QAA18862>