On Mon, 21 Dec 1998, John Forest wrote: > Tim Thomson wrote: > > I have mgetty running on my modem from inittab and it answers after 10 > > rings, but I've set up xringd to dialup the internet and then run mgetty. > > The second one complains about the first, and won't run. > > > > I would take the mgetty out of the inittab, but then when the line is used > > for ppp, it isn't logged in utmp, and I like this, because "sac dialout" > > gives the total time for the month. Is there any other way to get such a > > simple time log for dialup?
... > I think I understand what you mean. I set up mgetty and diald is now showing > up in 'last'. > Havent' tried yet, but maybe it is possible to set mgetty to debug 9 and make > up a script simulating xringd. I might be totally of the mark, but with a > little timing that program/script could check for RING's in the logfile. > Providing mgetty sends the RING's to the logfile of course. > If that works, you don't have to run xringd and 'sac dialout' would work. Hmmm, I haven't attempted this. What I did do was set up and inittab file with mgetty and one without, and put them in /etc/inittab.mg and /etc/inittab.nomg. Then I created /etc/mail.get which is an executable script like this : #! /bin/sh echo 'Dialing...' pppd call kapiti sleep 60 echo 'Waiting till pppd off...' while test -f /var/lock/LCK..ttyS3; do sleep 10; done sleep 5 echo 'Switching off mgetty via inittab' cp /etc/inittab.nomg /etc/inittab kill -HUP 1 sleep 2 echo 'Waiting for call...' mgetty -n 2 ttyS3 echo 'Modem Off, now turning mgetty back on' sleep 2 cp /etc/inittab.mg /etc/inittab kill -HUP 1 echo 'Done' ------------- I have had a problem with this though. After dialling up, and getting mail and stuff, it decided to shut down. But this is what happened: Dec 25 17:59:20 tui pppd[31393]: Terminating connection due to lack of activity. Dec 25 17:59:28 tui pppd[31393]: write: Interrupted system call(4) Dec 25 21:58:37 tui pppd[31393]: Hangup (SIGHUP) Dec 25 21:58:38 tui pppd[31393]: Exit. There is usually a "connection terminated" line in there, so I guess that part may have been the one to be interrupted. I guess pppd thought it was down, so wouldn't try again. It was up for 4 hours (at NZ$2/hr), before hanging up properly!!! I'm not sure if it was pppd finally timing out, or if my ISP did it automatically. I guess xringd or mgetty may have caused the interrupted system call, but I'm not sure how to fix it. I made the "sleep"s longer in /etc/mail.get to what the are now, but it still does it. It doesn't do this every time, usually only when there is a bit of traffic waiting. Any ideas? Thanks, Tim. -- ICQ #18073712 http://members.xoom.com/thomson
