It would be useful to have multiple servers listed.
Here is a machine, over in Australia.
ntpq> peers remote refid st t when poll reach delay offset jitter ============================================================================== *europium.canoni 193.79.237.14 2 u 49 64 377 331.811 -207.01 36.624
The impact of the significant delay and jitter is that ntp will prefer to step the time rather than slew it.
This means that I get dovecot (and other things) complaining about time moving backwards (http://wiki.dovecot.org/TimeMovedBackwards) fairly regularly.
It would be useful if Canonical could have a second, or third, machine in a different data centre also configured to provide NTP service and then have the default configuration updated to reduce the occurance of this problem.
Thanks, Anand
It would be useful to have multiple servers listed.
Here is a machine, over in Australia.
ntpq> peers ======= ======= ======= ======= ======= ======= ======= ======= ======= ======= =
remote refid st t when poll reach delay offset jitter
=======
*europium.canoni 193.79.237.14 2 u 49 64 377 331.811 -207.01 36.624
The impact of the significant delay and jitter is that ntp will prefer to step the time rather than slew it.
This means that I get dovecot (and other things) complaining about time moving backwards (http:// wiki.dovecot. org/TimeMovedBa ckwards) fairly regularly.
It would be useful if Canonical could have a second, or third, machine in a different data centre also configured to provide NTP service and then have the default configuration updated to reduce the occurance of this problem.
Thanks,
Anand