AutoPPP connection problems

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

I'm currently experiencing several different problems getting Win98
clients to connect
to this (Debian 2.1) box with mgetty-1.1.18.

I have had all working perfectly before, but unfortunately several
things have changed
at once.

Prior working setup was slackware w. mgetty-1.0.0, using a USR Sportster
Flash (un-flashed).

Now I've upgraded the server, started using Debian instead of slackware
(although that in
itself won't change anything, it does make it a little more difficult
for me to try
different versions of mgetty, or different compilation options), flashed
the sportster,
added another modem (MRI 56k external - rockwell chipset I believe), and
the telephone
switch has also changed.

First of all, the simplest problem. Calls coming in on the sportster
seem to have
intermittent problems with mgetty's call-type identification. It gets
the "CONNECT", but
then the next step screws up:

05/13 11:15:53 yS2  wfr: waiting for ``RING''
05/13 11:15:53 yS2   got: [0a][0d][0a]RING[0d]
05/13 11:15:53 yS2    CND: RING
05/13 11:15:53 yS2   wfr: rc=0, drn=0
05/13 11:15:53 yS2    CND: check no: 'none'
05/13 11:15:53 yS2  send: ATA[0d]
05/13 11:15:53 yS2  waiting for ``CONNECT''
05/13 11:15:53 yS2   got: ATA[0d]
05/13 11:15:53 yS2    CND: OKATA[0d][0a]+FDM[0d]
05/13 11:16:05 yS2    CND: +FDM[0a][0d][0a]CONNECT ** found **
05/13 11:16:09 yS2  send: 
05/13 11:16:09 yS2  waiting for ``_''
05/13 11:16:09 yS2   got: [0d]
05/13 11:16:09 yS2    CND:
CONNECT[1a][f0][f8][fe][fe][e6][e6]`[0c][00][06][0c][
e6]f[c0][cc][fe]f[c0][fc][e6][06][c0][cc][fe][e6][1e]3[86]`[cc][ff][86][18]3[06]
[c0][cc][fe][86]f[cc][ff][06][c0][fc][e6][06][c0][cc][fe]f[c3][cc][fe][86][18]3[
06][c0][cc][80]f[ff]f[cf][fc][00][9e][fe][e6][18]3[86]`[cc][ff][06][cc][cc][fe][
86]`[cc][ff]f[cf][cc][fe][e6]`[cc][ff][86][18][f3][9e][fe]

...and so on until it gives up. A successful connection on that modem
looks like this:

05/13 11:20:55 yS2  wfr: waiting for ``RING''
05/13 11:20:55 yS2   got: [0a][0d][0a]RING[0d]
05/13 11:20:55 yS2    CND: RING
05/13 11:20:55 yS2   wfr: rc=0, drn=0
05/13 11:20:55 yS2    CND: check no: 'none'
05/13 11:20:55 yS2  send: ATA[0d]
05/13 11:20:55 yS2  waiting for ``CONNECT''
05/13 11:20:55 yS2   got: ATA[0d]
05/13 11:20:55 yS2    CND: OKATA[0d][0a]+FDM[0d]
05/13 11:21:09 yS2    CND: +FDM[0a][0d][0a]CONNECT ** found **
05/13 11:21:10 yS2  send: 
05/13 11:21:10 yS2  waiting for ``_''
05/13 11:21:10 yS2   got: [0d]
05/13 11:21:10 yS2    CND: CONNECT[0a] ** found **
05/13 11:21:10 yS2   waiting for line to clear (VTIME), read:
~[ff]}#[c0]!}!}!} 
}7}"}&} }*} } }%}&} }'[c0][d6]}'}"}(}"}-}#}&[dd][90]~
05/13 11:21:10 yS2    looking for utmp entry... (my PID: 15767)


If I use minicom to dial into the sportster using the MRI, all the text
(login prompt etc.)
appears perfectly. If I dial into the MRI using the sportster, it loses
a chunk of the
welcome message right after the "CONNECT". I guess that the new
sportster firmware has
broken its handshaking in some way. Ugh.



The problem with the MRI is that it sometimes (intermittently) thinks
that the calling
modem is a fax. Doh.



If anyone else has experienced either of these problems, or can suggest
solutions, I'd
be very grateful.



Cheers,


Nick
--
nwp@blue-sky.co.uk