Date: Sun, 28 Jan 1996 15:57:11 -0700 (MST) From: Terry Lambert <terry@lambert.org> To: hsu@clinet.fi (Heikki Suonsivu) Cc: hm@altona.hamburg.com, freebsd-hackers@freebsd.org Subject: Re: Multi-Port Async Cards Message-ID: <199601282257.PAA01785@phaeton.artisoft.com> In-Reply-To: <199601280246.EAA23330@newzetor.clinet.fi> from "Heikki Suonsivu" at Jan 28, 96 04:46:46 am
next in thread | previous in thread | raw e-mail | index | archive | help
> mgetty would be useful as it can > actively reply calls instead of modems answering automatically, thus > avoiding callers getting modem answer when the terminal server has crashed > and won't be there. A properly configured modem and getty will do this as well. A properly configured modem will not answer unless DTR is asserted. DTR is asserted by the first open on a port (ie: getty). When the session is completed and a HUP delivered to all processes in the process group, the top level shell goes away, and without a group leader, the tty is revoked. So any process that listens to SIGHUP is dead and any process which does not is now in the background (there is some weird stuff with /dev/tty that I don't want the get into here). When the open count coes from 1->0, the DTR is dropped. This causes a correctly configured modem to reset (reset on on-to-off transition of DTR). Init restarts the getty, and the open is delayed by a proper driver so if the getty comes back too rapidly, the DTR is still low long enough for the modem to reset (usually, the delay period is >= 250mS, per Bell 103C signalling). When a modem resets, it drops the call (if any). If the machine crashes, getty goes away (the POST state on a proper serial port is to not assert DTR), and so the modems are not answered. 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?199601282257.PAA01785>