>From: callum@frost.bain.oz.au (Callum Gibson)
>Date: Thu, 19 Jan 1995 21:27:09 +1100 (EST)
>
>Has anyone had any success running term when logging in via kermit?
>I have only a 2400 baud modem at the moment so it needs all the help
>it can get! :-)  (probably a large hammer would help)  Anyway, I'm
>running standard 1.0-i386 release.  Under 0.9 I could use tip then ~^Z
>that and run term (all as root, mind you - but it did work).  Under
>1.0 I've had to get kermit use the modem properly as tip doesn't seem
>interested in it (always link down).  I have "local" in /etc/ttys as
>directed by others in the list, and in fact the kermit login works
>perfectly - I'm using it now.  But when I start up term, it gets wedged
>and starts timing out on packets pretty quickly. I've upped the timeout
>but I suspect there's something more sinister going on.  Also, when I
>manage to kill off the term/trsh the kermit is stuck and can't be killed
>(tries to exit but never succeeds - hangs on closing the device maybe?).

I recall seeing behavior like this under 0.9+ (Kermit refusing to
exit) when I had inadvertently set it up to use XON/XOFF flow control
and then something on the remote end sent an XOFF (^S) to my i486 box.

Make sure that you are using the latest version [5A(190)] of C-Kermit,
and SET FLOW RTS.  It is possible for older versions to claim that
they're using RTS/CTS when they're not really doing so; check with
stty.  If you make Kermit with 'make netbsd', then you don't even need
to use the 'local' flag in /etc/ttys.

I'm surprised that no one (including me) has ever brought up the
following:  Why does the i386 com driver hang in its close routine
when ixany is on and it has received an XOFF character from the remote
end of the link?  Shouldn't calling comclose() force termination of
the link (including dropping DTR) regardless of whether the local end
had been throttled with an XOFF or not?
-- 
Mike Long                                         Mike.Long@Analog.com
VLSI Design Engineer                   PGP 2.6.2 public key signature:
Analog Devices, CPD Division          CCBF225E7D3F7ECB2C8F7ABB15D9BE7B
Norwood, MA 02062 USA                assert(*this!=opinionof(Analog));