Hi, On Thursday 04 June 2009 10.05:36 MontyRee wrote: > 1.I set "restrict default kod nomodify notrap nopeer noquery" at ntpd > server. > > and according to this site, > http://www.eecis.udel.edu/~mills/ntp/html/accopt.html#restrict "noquery" > means 'Deny ntpq and ntpq queries. Time service is not affected.'. but > clients can ntpq query, why?
No idea. Post your entire ntp.conf. What ntpd version, on what operating
system?
> 2. In my test ntpd is shutdown automatically if the time interval is too
> big, What is the maximal time interval that can accept without shutdown?
> and ntpd is alive automatically if the time was corrected?
You can start it with "ntpd -g" so it steps the time instead of quitting
after startup. Normally, the threshold is 1000s.
> 3. If the time of the system was changed to 2008-01-01 suddenly,
> When the time of the system will be corrected(after some hours or
> days?) or not?
You should probably just try this.
> 4. If the remote time servers are five or more, the client
> query randomly or in order of precedence or more fast servers often?
In a normal configuration, it always starts querying servers every 64s and
then gradually slows the interval to 1024s. It will only fail to do so if
the network or the quartz of the local motherboard is extremely bad so that
it can't stabilize the time.
But it does rank the servers and weigh them into its error calculation
depending on their quality: there is a primary time source, there are "good"
servers and "not so good" servers ('*', '+', '-' in "ntpq -p" output) which
are used for the calculation of the correct local time. And then there are
falsetickers ('x') which are still queried but the time they return is not
used at all.
cheers
-- vbi
--
featured link: http://www.pool.ntp.org
signature.asc
Description: This is a digitally signed message part.
_______________________________________________ timekeepers mailing list [email protected] https://fortytwo.ch/mailman/cgi-bin/listinfo/timekeepers
