On May 26, 2011, at 10:48 AM, Miroslav Lichvar wrote: >> Evidently yes for your case. With only two servers, it may not be >> possible to find a best intersection via ntpd's variant of >> Marzullo's algorithm: > > I think the frequent source switching can happen with any number of > sources if the two best servers have similar synchronization distance.
Yes, I agree that it _could_ happen with any number of sources. > With more servers you may have a better chance that the best server is > significantly better than the others though. ...but, if you have a decent number of timeservers, the stability seems to improve significantly, and a change to the server rated as best seems to happen more in response to legitimate issues like a network problem somewhere, a timeserver being bounced, etc. > The problem can be usually fixed by increasing the anti-clockhopping > distance by "tos mindist", perhaps to 10 ms. This might be a useful suggestion for Florin, but I don't recall running into clockhopping issues with any great frequency even when I was running the older ntp-4.2.2 versions some years ago. Regards, -- -Chuck _______________________________________________ questions mailing list questions@lists.ntp.org http://lists.ntp.org/listinfo/questions
