🚀 go-pugleaf

In article <BxBGB4.84w@hsi.com>, tomg@code3 (Tom Green) writes: 
> We are trying to get hardware flow control going on an AOS/VS II system and 
> are having some trouble.  We set the HIFC and HOFC characteristics on the 
> con we are using, but it doesn't seem to be using regular rtc/cts signalling.
> We are trying to talk to a SCO unix box.  Anyone have any suggestions which 
> pins are really being used, or if we are off base totally in the first place? 

Sorry, I couldn't reply by email. 

Possibly pin 20* (and 6&8) are changing (in some ways this is closer to what
the original RS232 standard expected). But the best idea is to put a break-out
box with LED indicators between the two. Watch not only to see what handshaking
signals change, but whether something is sending an extra byte (XOFF). Is the
problem that the AOS system gets overloaded with input, or the SCO? - Or some
other problem? I had a printer on a DG box which kept getting too many
characters for its input buffer - the problem turned out to be the output
stopped from the DG a few characters after the printer said "please stop". The
moral of the story is that many, many things can be wrong with serial
communications. Expect a re-write of Murphy's Law real soon now, concentrating
on RS232!

* pin numbers refer to a DB25, not what comes out of the side of the mux.

Mark Aitchison, University of Canterbury, New Zealand.
(It's a while since I used AOS/VS - I upgraded to RDOS years ago :-).