** Reply to note from Kevin Saenz <[EMAIL PROTECTED]> Tue, 04 Nov 2003 12:24:07 +1100
>   
> why are you tuesday 10pm?

Kevin, that's a good Q.

the answer is long and involved, and, I do not understand some parts of
it...

but if you really need to know....:

this machine used to run 'daytime' against my web server, and, the
webserver run daytime against adelaide uni time server;

(not that corrections amounted to more than a second or two)

now that this machine is on intermittent dialup, I set the dialup to invoke
daytime augean.eleceng.adelaide.edu.au 
unknown to me, that no longer seems to be available
as a result of that, I was 1 hour out (DST)

so, today, when I noticed I'm out by DST, and, adelaide no longer is there,
I though, I'd try an NTPD instead of daytime, I've set up NTPD sometime in
1999, but, never used it since then, NTPD had these in:
poll interval = 16384
augean.eleceng.adelaide.edu.au
ntp.cs.mu.OZ.AU
ntp.ml.csiro.au
ntp.tip.CSIRO.AU
tick.usno.navy.mil
tock.usno.navy.mil
time.nist.gov
206.54.0.21

half of these no longer work, as I soon discovered, then, NTPD told me it's
correcting my clock by a weired amount of seconds.... and, I eneded really
way off. I then restarted NTPD again, again, it corrected my clock by some
weired amount. at which point I run daytime against tick/tock, and, it
fixed the time.

I guess, NTPD takes an average between local machine time as well as remote
clocks, and, I guess, NTPD shouldn't be invoked on on obviously incorrect
time, and, I guess, if I left NTPD running, it would eventually correct
the time. Perhaps an interval of '16384' prevented re-calc from being
somewhat quicker...

which reminds me, I should really configure ntpd on my Linux server.



Voytek Eymont
-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

Reply via email to