"ed" <[EMAIL PROTECTED]> wrote in message news:[EMAIL PROTECTED]
>> You do one or the other. Ntpd wants the clock for itself. Ntpdate >> doesn't care, but it's not as good as ntpd; it can't be - that would >> require having the clock to itself. > > I couldn't find the impact between ntpd and ntpdate from your word. > Ntpd will care the clock, ntpdate will not care. But what that > intersection of them? Or, ntpdate will change something which ntpd > will also use ? Yes. The clock. Ntpdate only takes snapshots, ntpd looks how the offset develops in trying to reduce it to zero and keeping it at zero. Ntpd does things like observing the offset a known interval apart. If the interval isn't reliable (because ntpdate has changed the clock in the meantime), ntpd can't estimate trends. Groetjes, Maarten Wiltink _______________________________________________ questions mailing list [email protected] https://lists.ntp.isc.org/mailman/listinfo/questions
