🚀 go-pugleaf

Bruce Momjian (root@candle.uucp) wrote:
: Jon LaBadie (jon@jonlab.UUCP) wrote:
: : Having used a clock synchronization program on my 3B1 computer,
: : I was pleased to see settime.c come across comp.sources.misc.
: : 
: 
: I hacked up the source to use 'cu' for System V machines.  Let me know
: if anyone wants a copy.  It is not as accurate because all data passes
: through a pipe from cu to the main program.

My clock is so drifty, the error introduced by the pipe delay
is insignificant.

You can look for comp.sources.misc Volume 3, Issue 79 "utc"
that can use cu.  It can also serve as a time source,
so you can get the time via a cron cu|utc job every
day or so and share it with others
calling your system.

In old XENIX systems, stime() did not set the AT cmos
clock, so I wrote 'cmostime3' that did it all:
acquire a tty and lock it, dial the numebr, scan for the time,
hang up, stime() and write /dev/cmos or /dev/rtc.

I thought this went out to comp.sources.misc in 1990 but I
cannot find it in the archive.  if there is any interest,
I can make it (or a later rev cmostime4) available.

cmostime is a bit overbearing, even obsequious, but it does
work and it depends on very, very little other than termio
and stime() :-).  It is a bit SCO specific, looking
to write a lock file in /usr/spool/uucp, not /usr/spool/locks.

Dial(3) has never worked properly anywhere I have encountered it.
Most places I have found it, it didn't even know HDB had
replaced version 2 :-|.  Evidently, the 3B folks got it right.

------------------------------------------------------------------------------
Warren Tucker  (404)587-5766              n4hgf!wht or wht@n4hgf.Mt-Park.GA.US
All readers cannot be leaders, but all leaders must be readers. - Harry Truman
-- 
------------------------------------------------------------------------------
Warren Tucker  (404)587-5766              n4hgf!wht or wht@n4hgf.Mt-Park.GA.US
All readers cannot be leaders, but all leaders must be readers. - Harry Truman