🚀 go-pugleaf

I'd like to thank all those who responded to my previous query
about getting PPP from SCO TCP/IP 1.2.0i to work. The consensus
was (all 8 replies sofar) that I either look into a newer version
(1.2.1) from SCO or get the PPP package from MorningStar. The current
package just doesn't work and thanks to the replies from the net
I avoided wasting more time on it (I spent two whole days with SCO
PPP 1.2 before asking).

I should add that MorningStar PPP works well on SunOS and I am
considering getting it for SCO as well. Anyway, I wonder if SCO
issued an SLS for PPP yet ...

I encountered the following problem using SCO's SLIP (again from
SCO TCPIP 1.2.0i -- slattach): I had a Mac connected over a null-
modem with hardware flow control on the port. I switched off the
Mac for the night (the RS232 tester showed that CD dropped) and
next morning the SCO system was unusable: the console accepted no
commands and logins from other ports all failed. The error message
on the console was:

CONFIG: getcf - Out of clists (NCLIST=1000 exceeded) xxx failures sofar.

The system was obviously out of clists, although I had raised them
from the default of 120 to 1000 following an advice from FAS, an excellent
asy/sio replacement program, just to avoid such a situation. I
suspected that it had to do with the hardware flow control configuration
of the port, which resulted in the SLIP packets being accumulated
and thus eating all available clists. So I turned on the Mac, got into
a terminal program which asserted HFC and CD and then received all the
SLIP packets which were accumulated during the night. The problem was
temprarily solved.

Since I'd like to run SCO's SLIP over leased lines, I need to make it
recognize the drop of the carrier or do a timeout on the port, so that
it won't hang everytime the leased line is unavailable for a few hours.

If someone solved this problem, please let me know. I'll summarize and
report to the net.

Best regards,

Dimitris.