[ntp:questions] Synchronizing against a free-running server

David Woolley david at ex.djwhome.demon.invalid
Mon Apr 15 12:22:57 UTC 2013


detha wrote:

> rootdispersion=10802.399, refid=LOCL, reach=377, unreach=0, hmode=3,
                  ^^^^^^^^^
> pmode=4, hpoll=10, ppoll=10, flash=400 peer_dist, keyid=0, ttl=0,
                                          ^^^^^^^^^

Dispersion is too high because it is too long since the server 
synchronised to any valid source.  Faking root dispersion is the real 
reason that one runs local clock drivers on free-running reference 
implementation servers.

I don't know if and how w32time can be forced to lie about the root 
dispersion.

You may be able to set tos maxdist on the clients, assuming they are 
sufficiently recent, to make it tolerate root distances of more than 1.5 
seconds (you'd need 11 seconds at the instant at which you ran rv, but 
this value will grow without bounds).


> 
> It's what w32time seems to do (I don't know too much about windows, so
> maybe that can be tweaked)


Recent versions of w32time can be tweaked to be NTP compliant, or at 
least more or less so, but supporting a local clock driver is not part 
or the requirements of the standard.



More information about the questions mailing list