[ntp:questions] What happens if ntp server unavailable at start up?

Martin Burnicki martin.burnicki at meinberg.de
Tue Sep 16 07:40:59 UTC 2008

Unruh wrote:
> Martin Burnicki <martin.burnicki at meinberg.de> writes:
>>Unruh wrote:
>>> Steve Kostecke <kostecke at ntp.org> writes:
>>>>On 2008-09-13, Joseph Gwinn <joegwinn at comcast.net> wrote:
>>>>> Unruh <unruh-spam at physics.ubc.ca> wrote:
>>>>>> Harlan Stenn <stenn at ntp.org> writes:
>>>>>> >Unruh> Did the dynamic keyword ever work? The web docs say that it
>>>>>> >Unruh> is not yet implimented.
>>As already mentioned earlier, the keywaord "dynamic" had been introduced
>>IIRC in 4.2.4. It lets ntpd retry if a server can not be reached
>>However, since the behaviour with the "dynamic" keyword is what users
>>would normally expect ntpd to do I have proposed to make the "dynamic"
>>behaviour the default and obsolete the "dynamic" keyword. Frank Kardel who
>>had introduced the "dynamic" keyword agreed and implemented the proposed
>>changes in ntp-dev 4.2.5p38 on 2007-05-26, and he also removed the
>>description of the keyword from the HTML docs.
>>Unfortunately Dave Mills who maintains the HTML docs at udel.edu did not
>>pick up Frank's changes of the HTML docs. The file history for
>>confopt.html shows that the description of the "dynamic" keyword has been
>>re-added when confopt.html changes from Dave were pulled in again on
>>I did a backport of Frank's changes for ntp-stable and I also modified the
>>HTML docs and removed the description of the "dynamic" keyword. Those
>>changes were picked up in 4.2.4p5.
>>So if you use the stable version and have a look at the HTML docs which
>>come with that version this should be OK.
> I looked at the docs on the official site.

As mentioned frequently on this page, the docs at www.ntp.org only refer to
the ntp-dev version (in this case the online docs are not correct, though).

>>>>If you wish to use Google to search the on-line stable release
>>>>documentation use "site:doc.ntp.org" in your search arguments.
>>> Yes, I went there to search on the word dynamic. I got junk.
>>Hm, I also tried this (searched for "dynamic site:doc.ntp.org") and got a
>>couple of hits for 4.2.4, where the "dynamic" keyword is described for
>>ntpdc and as "server option" in confopt.html.
>>Since this refers to 4.2.4 but not 4.2.4p5 where it has been obsoleted
>>this is absolutely correct, and I don't know why you would consider this
>>as junk.
> The "junk" was irelevant hits ( about 20 of them to some refclock MX4200
> receiver, one to ntpc, The only relevant hit was one with a "relevance
> factor" or 1 star which was something like the 17th hit out of 23.
> That is what I call "junk". It often happens when googling I agree.
> Note I used the search on the doc.ntp.org main page, not a google search.

How should a search engine know whether you are searching for the "dynamic"
keyword or for the "Dynamic, 3D Nav" mode of an MX4200 refclock?

If you select boolean search mode on docs.ntp.org and enter 
"dynamic and option" 
then there are much better hits. 

BTW, the search result of the ntpdc docs is correct since ntpdc also had to
deal with the "dynamic" keyword.

Martin Burnicki

Meinberg Funkuhren
Bad Pyrmont

More information about the questions mailing list