mgetty swallows chars ?

• Messages sorted by:[ date ][ thread ][ subject ][ author ]

On Wed, 14 May 1997, Gert Doering wrote:

> > mgetty is set up, an incoming call is answered,
> > and the caller sees somewhat like this on the screen:

> > Welx 2.0.30 !

> > The first line should normally be "Welcome to Linux 2.0.30 !",
> > the string "come to Linu" is missing, although the file
> > /etc/issue contains the line correctly.

> Never seen something like this. What speed are you using on the modem-PC
> serial cable? 

I have set it up to 115.2 kBit/s. But it doesn't matter how high
the speed is, it's always there.

> Maybe the cable is too long for that speed, or not good
> enough.

Well, I've tried another two cables, and again, the same
behaviour.

> > If I use my Zyxel U-1496 EG+, all works well.

> I assume that you are accessing the 1496 slower?

Jep, at 57.6 kBit/s.

> > Mgetty works, with the above-mentioned exception, very well
> > with the Elite 2864, no errors, no problems.

> This is interesting. If it's a cable problem, you should see problems 
> during the connection as well (for example, file transfers should be
> slower than usual, because characters get lost).

No, there are no significant problems or errors during the 
connection /links.
The cps-rates are good, also.

But you have animated me to try out the hardware, and I found
the solution of this problem, finally ;)

It is the onboard-highspeed I/O of the ASUS P55TP4N mainboard !
It is *not defective* (I'm sure), but somehow buggy, I have build in a
little I/O card with an original 16C550 AF, and now 
it works like it should.

This is also the answer to the question why the U-1496 has not
shown this strange behaviour, too, because I connected it,
for testing purposes, to an extra built-in serial card.

Tomorrow I will test if the U-1496 also shows this behaviour 
together with the ASUS onboard-I/O (just to know if).

> It's not a known bug, and unfortunately, I do not know a solution either
> -- my 2864DI does *not* do that (call 089/35731018 w/ X.75 and convince
> yourself...) -- nothing special in the setup. Short RS232 cable, 115200
> bps.

Well, in fact, now, I do believe that :)

I've tested another ASUS P55TP4N mainboard, and it's the same !
The multi I/O on this mainboard (a great mainboard, I think),
shows also the known behaviour. Maybe it's because it uses
the SMC controller chip, of which I read one year ago that it
has some bugs ....

Thank you for the hint with the cable, of without it I wouldn't
have thought about the serial hardware as the cause (at least not during
the next days, because it seemed to work alright since over one year,
uucico shows constant transfer rates up-to 7550 cps).

Regards, Heinz.
-- 
# Heinz Diehl, 68259 Mannheim, Germany
# PGP-encrypted mails welcome, key on request 
# "Tomorrow never knows what it doesn't know too soon" - Oasis

.