mgetty problem remains...
Gert Doering (gert@greenie.muc.de)
Wed, 29 Jan 1997 10:52:49 +0100
Hi,
Martin Lorenz wrote:
> i am experiencing problems with mgetty:
> it is not possible to dial in and as long as mgetty is running i canīt
> ppp out! i have to work around this by changing my initab each time i
> want to dial out (faxrunq works well on the same port...)
This is very straightforward: you have the ppp locking wrong. Make sure
pppd will create the lock file mgetty is looking for (/var/log/LCK..ttyS2)
*before* it touches the serial device. Most likely, you only need to
specify the "lock" option to pppd.
> i am attaching parts of my messages and log_mg.ttyS2 files
> maybe some of you know whatīs happening here... i donīt
It's very clear from the log_mg.ttyS2 file.
> 01/28 05:14:03 yS2 removing lock file
> 01/28 05:14:03 yS2 waiting...
mgetty waiting for "activity".
> 01/28 05:25:01 yS2 select returned 1
There was some activity on the serial port. Is it a dial-in or dial-out?
> 01/28 05:25:01 yS2 checking lockfiles, locking the line
> 01/28 05:25:01 yS2 makelock(ttyS2) called
> 01/28 05:25:01 yS2 do_makelock: lock='/var/lock/LCK..ttyS2'
> 01/28 05:25:01 yS2 lock made
There is no lock file owned by "someone else", so either the dial-out
process goofed, or it's a dial-in. Mgetty locks the device, to make sure
all dial-out processes know the port is in use.
> 01/28 05:25:01 yS2 tcgetattr failed: I/O error
> 01/28 05:25:01 yS2 tcsetattr failed: I/O error
This should never happen, so it's likely the dialout process is pppd and
has changed the line discipline.
> 01/28 05:25:01 yS2 waiting for ``RING''
> 01/28 05:25:01 yS2 got:
> 01/28 05:25:18 yS2 timeout in chat script, waiting for `RING'
Waiting for RING will - naturally - fail as there is data going out, not
coming in. And so on.
gert
--
USENET is *not* the non-clickable part of WWW!
//www.muc.de/~gert/
Gert Doering - Munich, Germany gert@greenie.muc.de
fax: +49-89-3545980 gert.doering@physik.tu-muenchen.de