On Fri, Feb 29, 2008 at 11:16:13PM -0700, Darrin Chandler wrote: > On Fri, Feb 29, 2008 at 09:54:00PM -0600, Michael Rathbun wrote: > > On Fri, 29 Feb 2008 20:25:26 -0700, Darrin Chandler > > <[EMAIL PROTECTED]> wrote: > > > > >Sarcasm aside, and baiting aside, is there anything about these servers > > >that makes them unsuitable for pool servers? I'm sure there are nits and > > >minor criticisms, valid and otherwise. But if we keep to the real issue > > >about the pool then I'd be happy to know why these servers don't fit the > > >bill. > > > > Well, there's apparently a superior peak-to-average ratio when compared > > to time.rabendary.net (<http://www.pool.ntp.org/scores/64.183.187.9>), > > which we really can't have, old boy. > > I've put one of my two servers on xntpd, and both now share the same set > for upstream. It'll be interesting to see how it looks after it settles > in.
I spent some quality time looking at upstream servers and settled on some with low latency to my servers. Once everything settled in they started to look very good, but then the monitoring issues began. Now that things have been back to normal for a while both servers are doing quite well, and are very comparable in terms of quality. One of them is using OpenNTPD, the other running the ntp.org ntpd. http://www.pool.ntp.org/scores/140.99.51.114 http://www.pool.ntp.org/scores/140.99.51.115 So it seems like picking good servers is MUCH more important than whether or not OpenNTPD is involved. If any of you care to take another look at these servers and spot any more problems with OpenNTPD I would be glad for the information. And thanks to all who spent time looking at this before. -- Darrin Chandler | Phoenix BSD User Group | MetaBUG [EMAIL PROTECTED] | http://phxbug.org/ | http://metabug.org/ http://www.stilyagin.com/ | Daemons in the Desert | Global BUG Federation _______________________________________________ timekeepers mailing list [email protected] https://fortytwo.ch/mailman/cgi-bin/listinfo/timekeepers
