On Apr 6, 12:32 pm, Steve Kostecke <koste...@ntp.org> wrote: > On 2009-04-06, David Woolley <da...@djwhome.demon.co.uk> wrote: > > > Seems to me that iforphanmodeis using narrow error bands, it is not > > going to work in many or most cases where people would want to use it! > > > It needs to work when there never was a real time source (probably the > > main reason people will actually use it), and all the machines are > > started together with their clocks randomly spread over at least 10 > > seconds. Otherewise, it is not going to meet typical, naive, user > > expectations. > > OrphanModeis intended to select _one_ leader from the group of Orphans > when all other time sources are unreachable. It does so based on a > randomly generated value. > > -- > Steve Kostecke <koste...@ntp.org> > NTP Public Services Project -http://support.ntp.org/
I read that NTP uses "rootdelay" and "rootdispersion" variables to select leader among themselves in SAME stratum. "randomly generated value" - Is it stored in a ntp variable that we can see ? _______________________________________________ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
