Mischanko, Edward T wrote:
How can my local clock have a negative offset when none of the
*** selected peers have a negative offset? Is it possible?
It is not clear what you mean by "local clock" and "offset", here. For the normal NTP meanings of these, the offset of the local clock is zero by definition.
Could you provide the ntpq, or other, output on which you are making this observation, and indicate which fields you are using.
In the normal ntpd sense, the offset of the local clock is always zero and the offsets of other sources should be randomly scattered around zero, if the time is properly synchronised (the distribution may not be symmetric, but can never be one sided, once the clock is locked.
Offset is measured relative to the estimated time measurement from the upstream server. It is not relative any absolute standard of time, except after adding all the intermediate errors.
_______________________________________________ questions mailing list [email protected] http://lists.ntp.org/listinfo/questions
