Chris Hastie wrote:
> On Tue, 7 Feb 2006, Jeffrey B. Ferland <[EMAIL PROTECTED]> wrote
>>
>> I've noticed that I stay very consistent in my offset, yet I see 
>> giant offset spikes in the chart, and noticed at one point more than 
>> 10 days back that about 1/2 the servers in the pool got dinged for 
>> offset in one particular check. My curiosity is how a server with 
>> peers like those below is seen by the mon server with spiking  offsets.
>
> Network issues. A big delay, different in each direction, will
> introduce errors in what the monitoring servers see. It doesn't mean
> the clock on the monitored box is wrong, and it doesn't necessarily
> mean that a different server will see the same offset.
>
> As an example, http://www.pool.ntp.org/scores/62.3.200.116 shows the
> offset regularly spiking at 40ms, sometimes touching 70ms.
>
> http://ntp.raggedstaff.net/offsets2.php shows the offsets measured
> from 62.3.200.116 to its peers. This suggests the clock is rarely more
> than around 3ms out.
>
> http://rupert.raggedstaff.net/status/offsets.php shows offsets
> measured from another machine on a remote network, but not nearly as
> remote as the monitoring server. 62.3.200.116 is shown as the trace
> marked ash.oak-wood.co.uk. Significant offsets are much less common
> seen from here than from the monitoring server, and rarely over 10ms.

I'm seeing similar spikes for my servers - one of which is stratum 1
varies by +/- 40us typically.   I put it down to the monitoring server
either having a really bad time source or an overloaded net connection.
- Where is the monitoring server and how is it connected?



john
_______________________________________________
timekeepers mailing list
[email protected]
https://fortytwo.ch/mailman/cgi-bin/listinfo/timekeepers

Reply via email to