[ntp:questions] Remaining synced on an unsynchronised peer?

Dave Hart davehart at gmail.com
Tue Dec 1 10:10:41 UTC 2009


On Nov 30 23:52 UTC, Michael Butow wrote:
> What I did observe is fluctuating reachability on them, alternating
> fairly regularly from one to the other.

In another message you mentioned seeing reach alternate between 376
and 377 on the symmetric ("peer") association.  That is normal for
symmetric mode and does not indicate a reachability issue.  Anything
other than 376 or 377 indicates at least one missing packet, once
primed.

> However, I feel there's got to be a bug when an unsynchronised peer
> continues providing replies.

ntpd which has reached 0 reachability for all its sources should be
responding with a root dispersion well over the 1s threshold and be
distrusted by downstream ntpds.  Assuming no orphan-mode
configuration, it is not clear to me if the stratum is supposed to go
the maximum and the synchronization (leap bits) to 11, or if they are
expected to remain at the last values with only the root dispersion to
indicate the problem.

I pick a nit about calling a ntpd unsynchronized as soon as its
sources are severed.  A "freewheeling" ntpd simply maintaining the
last known frequency compensation is initially still a good source of
time, decreasingly so over time.  NTP 4 maintains an error budget
along those lines, resulting in the root dispersion value that's part
of every NTP response.

Cheers,
Dave Hart




More information about the questions mailing list