Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.3 4.3bsd-beta 6/6/85; site spdcc.UUCP
Path: utzoo!watmath!clyde!burl!ulysses!mhuxr!mhuxt!houxm!ihnp4!spdcc!dyer
From: dyer@spdcc.UUCP (Steve Dyer)
Newsgroups: net.micro.pc,net.unix
Subject: SCO XENIX V problem
Message-ID: <26@spdcc.UUCP>
Date: Mon, 3-Mar-86 22:00:16 EST
Article-I.D.: spdcc.26
Posted: Mon Mar  3 22:00:16 1986
Date-Received: Wed, 5-Mar-86 04:01:28 EST
Organization: S.P. Dyer Computer Consulting, Cambridge MA
Lines: 24
Xref: watmath net.micro.pc:7266 net.unix:7287
Summary: getty begins to fail on serial dial-in line after 12 hours

I made a call in to SCO support last week, but I'm hoping that a
faster response may come from someone out there who has already seen
this and received an answer from SCO.

Briefly, I have two modems on my AT, the first (COM1: in DOS talk)
is /dev/tty11 and is used for dial-out.  The second, /dev/tty14, is
connected to an auto-answer modem and is used for incoming calls.
tty14 is entered in /etc/ttys as enabled and is on a 1200-2400-300
baud rate cycle.

After about 12 hours of operation, getty begins to fail when an
incoming call arrives: upon receipt of carrier, it reports:
cannot open tty14: error 25 (which happens to mean "not a typewriter.")
Rebooting is the only option at this point, at least if I want to
make tty14 available again for incoming calls.  From this side of the
source license, it looks like getty is sleeping on carrier, but when it
finally arrives, errno 25 is generated somehow, and the condition never
clears.

Any clues would be appreciated.
-- 
Steve Dyer
dyer@harvard.HARVARD.EDU
{bbncca,bbnccv,harvard}!spdcc!dyer