[I think this is a Linux rather than a UUCP question. :-)]
I have been "playing" the UUCP capability of LInux SLS 1.03 for the
past few days (and nights. :-)) I used /dev/modem which is softlinked
to /dev/cua1 (COM 2) with uugetty -r set for /dev/modem in
/etc/inittab. Also I used Any in the relevent entries in the
/usr/lib/uucp/Systems file (just to test things out).
Here are several observations:
1) When SLS 1.03 is first installed, the system talks to the modem
(but no dialing out) frequently and then after a while (like an hour
or more), it leaves the modem alone.
2) Having done the necessary modifications to Devices, Dialers, and
Systems in the /usr/lib/uucp directory and the paths and config
files in the smail directory (don't remember the exact path
anymore), I could test my UUCP link using the command:
/usr/lib/uucico [-f] -x2 -s<hostname>
Without the -x2 (or higher e.g. x3, x4) flag, Linux would not
initiate a call otherwise uucico dials out with and without the -f
or the -r flags set.
3) I use crontab to schedule dialouts with the -f and -x2 flag on (I
want to make sure that the system dials out). Everything works
fine for a day or so and then
/usr/lib/uucico -f -r1 -x2 -s<hostname>
would do nothing except create the LCK..* and TMP...... files
in /usr/spool/uucp. The audit.local and other logs in the
/usr/spool/uucp don't seem to get updated when this happens (i.e.,
when uucico doesn't initiate a dialout). Usually the logs will say
something or explain why the dialout didn't occur and a couple of
modem lights will flicker for a few seconds but in this case nothing.
a) How tightly "hardcoded" is the uucp neighbor "quick" to SLS 1.03?
b) How fast does the C.* and D.* files get created in the
/usr/spool/uucp/hostname directory? I notice that sometimes those
files (email addressed to hostname) get created as soon as I "send"
them out of the mail program (Emacs rmail mode) but then there are times
I notice that the files don't get created that quickly.
c) I haven't left my system alone long enough (e.g. for > 24 hours)
to see if the UUCP link will "wake" up again. Is the uucico's behavior
described in 3 above expected?