Fwd: Using ntpdate -b SERVER shortly after SERVER boots

Fwd: Using ntpdate -b SERVER shortly after SERVER boots

Post by donaldm31 » Wed, 14 Feb 2007 04:46:21


Oops. Sorry for the off-list reply.

---------- Forwarded message ----------
From: Donald Murray, P.Eng. < XXXX@XXXXX.COM >
Date: Feb 12, 2007 12:45 PM
Subject: Re: Using ntpdate -b SERVER shortly after SERVER boots
To: Steve Kostecke < XXXX@XXXXX.COM >




Trying to be clueful here. I'm a big proponent of ntp, and long overdue
in joining this list.


For our systems it's simpler to use the ntpdate/date approach than
start ntpd and wait for it to sync. We need to know when the clock
may be stepped, and insure our apps are not running.


It's a CGI on the SERVER, called by a script on the CLIENT.


Yes, we then start ntpd.

Agreed. Glad I'm not the only one that thinks this is reasonable
given our unreasonable constraints.

Thanks again!
_______________________________________________
questions mailing list
XXXX@XXXXX.COM
https://lists.ntp.isc.org/mailman/listinfo/questions
 
 
 

Fwd: Using ntpdate -b SERVER shortly after SERVER boots

Post by Harlan Ste » Wed, 14 Feb 2007 06:52:55

Donald,

Have you seen http://www.yqcomputer.com/

H

 
 
 

Fwd: Using ntpdate -b SERVER shortly after SERVER boots

Post by donaldm31 » Wed, 14 Feb 2007 07:12:36


Yes. Until I started looking into this issue I wasn't aware there of
this site at all. It's full of lots of useful information.

Was there anything in particular I should note on the StartingNTP4
page? I didn't see anything that I thought would more quickly get
SERVER's ntpd available to ntpdate queries on the CLIENT.
_______________________________________________
questions mailing list
XXXX@XXXXX.COM
https://lists.ntp.isc.org/mailman/listinfo/questions
 
 
 

Fwd: Using ntpdate -b SERVER shortly after SERVER boots

Post by Richard B. » Wed, 14 Feb 2007 08:05:29


That's an interesting article but it misuses the word "precision".
Precision is a property of your clock and can be thought of as the
smallest possible difference in time that the clock can measure.

A clock can have a precision of 1 microsecond and still be five minutes
slow!
 
 
 

Fwd: Using ntpdate -b SERVER shortly after SERVER boots

Post by sten » Wed, 14 Feb 2007 08:06:55

Aside from the other information you have learned, I was thinking about
the ntp-wait script, if that was an option.

I see two issues here, one has to do with getting your server to
announce the time ASAP, and the other is getting the clients to set
their clocks as soon as the server is ready.

ntp-wait will help with the 2nd part.

H
_______________________________________________
questions mailing list
XXXX@XXXXX.COM
https://lists.ntp.isc.org/mailman/listinfo/questions
 
 
 

Fwd: Using ntpdate -b SERVER shortly after SERVER boots

Post by Harlan Ste » Wed, 14 Feb 2007 14:59:45

>>> In article < XXXX@XXXXX.COM >, "Richard B. gilbert" < XXXX@XXXXX.COM > writes:
Harlan> Have you seen http://www.yqcomputer.com/

Richard> That's an interesting article but it misuses the word
Richard> "precision".

It's a twiki. Feel free to make improvements.

H
 
 
 

Fwd: Using ntpdate -b SERVER shortly after SERVER boots

Post by Uwe Klei » Wed, 14 Feb 2007 21:44:42


isn't that "resolution" ?
iananes ( i am not a native english speaker ;-)

uwe
 
 
 

Fwd: Using ntpdate -b SERVER shortly after SERVER boots

Post by David J Ta » Wed, 14 Feb 2007 22:27:15


[]

Uwe,

I suspect these terms mean what you choose them to mean. Here is what
Google has to offer:

http://www.yqcomputer.com/ %3a+precision

(includes both "number of digits" and "degree of refinement")

http://www.yqcomputer.com/ %3A+resolution&btnG=Search

(includes "measure of accuracy or dynamic range of an A/D or D/A
converter" and "smallest quantity that can be measured").

From this, I would suggest:

precision - degree of correctness

resolution - accuracy of measurement

Perhaps NTP has its own definitions?

Cheers,
David
 
 
 

Fwd: Using ntpdate -b SERVER shortly after SERVER boots

Post by Richard B. » Thu, 15 Feb 2007 01:55:55


Don't worry about it. Your English is SO much better than my Deutch.
(I can make a few polite noises in Espal, Dech, Nihongo (Japanese)
and Turkish). I'm fluent only in English and it could be argued
convincingly that I'm not fluent in English either!

Resolution is a good synonym but NTP usage describes the "precision" of
a clock in terms of the value of the least significant bit or the
smallest measurable interval. I think RFC-1305 defines precision
somewhere but it's hard to find without reading cover-to-cover. I hope
the V4 RFC has an index!!

<snip>
 
 
 

Fwd: Using ntpdate -b SERVER shortly after SERVER boots

Post by Richard B. » Thu, 15 Feb 2007 02:11:33


Indeed it does! See RFC-1305 page 11.
 
 
 

Fwd: Using ntpdate -b SERVER shortly after SERVER boots

Post by Uwe Klei » Thu, 15 Feb 2007 02:22:45


What about a glossary of terms used?

uwe
 
 
 

Fwd: Using ntpdate -b SERVER shortly after SERVER boots

Post by Richard B. » Thu, 15 Feb 2007 05:14:17


<snip>
<snip>

A glossary would help too. At 113 8-1/2x11" pages, it's a fair sized
book and both an index and a glossary would be appropriate additions.
 
 
 

Fwd: Using ntpdate -b SERVER shortly after SERVER boots

Post by Harlan Ste » Thu, 15 Feb 2007 09:28:06

>>> In article < XXXX@XXXXX.COM >, Uwe Klein < XXXX@XXXXX.COM > writes:

Uwe> What about a glossary of terms used?

http://www.yqcomputer.com/

H
 
 
 

Fwd: Using ntpdate -b SERVER shortly after SERVER boots

Post by Uwe Klei » Thu, 15 Feb 2007 17:33:19


Hey, Thanks.
No wishes remaining for Xmas.

uwe
 
 
 

Fwd: Using ntpdate -b SERVER shortly after SERVER boots

Post by mill » Fri, 16 Feb 2007 19:12:45

Richard,

The rfc1305 is four *** years old and now overtaken by the NTPv4 draft
now on the project page. I would be happy to add a glossary and index,
but not in Postel ASCII. My tools produce wonderful, beautiful
PostScript and PDF, but not Postel ASCII.

Dave