SUMMARY: ALM flow control problems

From: Daryl Crandall (daryl@oceanus.mitre.org)
Date: Mon Aug 19 1991 - 13:12:36 CDT


Sun-managers;

Here's the summary of my question about XON/XOFF flow control problems with
an ALM-1 on a Sun4/380 under SunOS-4.1.1:

        Daryl Crandall
        The Mitre Corporation
        daryl@mitre.org
        (703) 883-7278
        
############################################################################
SUMMARY:

The patch that I finally installed was the 'tty_ldterm.o' from
patch #100225-02 This seems to have solved my problem.

The patch #100225 is listed as fixing a problem with the ALM-II but I was
told by Sun help personel to apply only the 'tty_ldterm.o' since the flow
control problems were caused by a bug in the higher level software, not
down at the physical driver level.

I appologize for the delay but I wanted to see if this patch actually
fixed something and didn't introduce new problems. So far it seems to be OK.

Other serial I/O patches were mention but I won't confuse the issue by listing
them since I didn't test them.

During this process I came across the (unverified) information that the
ALM-1 board is unsupported on 4/3xx and 4/4xx machines but is still supported
on 4/2xx machines. (NOTE: "unsupported" doesn't necessarily mean "doesn't
work!")

Thanks to all who responded and especially:

tessi!joey@nosun.West.Sun.COM (Joe Pruett)
Mike.McCann@eng.clemson.edu (Mike McCann)

############################################################################
ORIGINAL QUESTION:

I've got an ALM (aka: Systek, MTI 800/1600 Multi-Terminal-Interface) serial
multiplexer board (NOTE: not an ALM-II, this is the earlier ALM).
This board is controlled by A Sun4/380 (upgraded 3/180) under SunOS-4.1.1
The upgrade simply swapped CPU boards to convert us to a Sun4/380.

All our printers are attached to this board at 9600 baud and use software
flow control (XON/XOFF). Since the upgrade, we've been experiencing
what appear to be flow control problems in which the four printers (Texas
Instruments TI-2115 and Apple LW-II/NTX) hang waiting for data from the
server. The TI-2115 actually says "Waiting" in it's display.

The printers worked fine for 3 years under the Sun3/180 using SunOS-4.0.3
and earlier. The problem is random and fixable by using lpc to "abort" then
"start" the line printer daemons.

I've given Sun (800 USA-4SUN) the chance to identify and fix this problem
for us but after initial promises of an E-mail patch (#100137-01) I had
to call back 4 days later to ask again. Now that the patch has arrived,
I notice that the README indicates that it is for Sun3 not Sun4. Grrrr.
The person handling the service call says "I'll call you back ...".

Meanwhile, this problem is a real pain since someone has to be available
more often than usual to immediately kick the daemons when a printer hangs.

Any suggestions?

Things, I've considered:

        experiment with hardware flow control (may need to run new cables)

        examine the Sun3 patch carefully to see if it might possibly work
        for a sun4 also. (doubtful and/or dangerous)

        Carefully examine kernal configuration parameters (GENERIC) for
        mti flag options set wrong.

############################################################################



This archive was generated by hypermail 2.1.2 : Fri Sep 28 2001 - 23:06:20 CDT