On 11-02-08 00:57, Ask Bjørn Hansen wrote:
>
> On Feb 10, 2008, at 2:09 PM, Arnold Schekkerman wrote:
>> 1202631052,"2008-02-10 08:10:52",-0.00433087348937988,-4,13.7 !!!
>
> If the stratum is 6 or higher then you get a negative score. I expected
> that to be high enough to not happen for otherwise good servers. I've
> changed it to 8 or higher.
>From my config I would say my system would be stratum 4 at most, but
according to my log it was even stratum 2 (sync'ed to the st 1 server) from
Feb 6 to Feb 11. The system is running for 37 days with a stable PLL
Frequency (61.400 +/- 0.030 for the past week).
My peers are:
remote refid st poll reach delay offset
==============================================================================
-194.109.22.18 129.132.2.21 3 1024 377 6.897 10.806
-131.211.80.155 131.211.80.22 2 1024 377 10.183 8.516
+204.9.53.11 64.183.55.54 2 1024 377 7.586 3.855
-82.95.238.2 193.79.237.14 2 1024 377 16.427 26.800
*213.84.46.114 .DCFa. 1 1024 377 15.114 -3.587
+145.24.129.6 192.87.36.4 2 1024 377 12.554 5.576
-217.114.97.98 130.88.202.49 3 1024 377 11.180 10.544
127.127.1.0 .LOCL. 10 64 377
>From my log files:
Feb 6 12:27:58 gateway ntpd[6090]: synchronized to 213.84.46.114, stratum 1
Feb 11 13:47:20 gateway ntpd[6090]: synchronized to 131.211.80.155, stratum 2
Feb 11 13:59:01 gateway ntpd[6090]: synchronized to 213.84.46.114, stratum 1
> I've mentioned it before: The monitoring system isn't trying to measure
> more accurately than 100-200ms (the point where the score starts being
> negatively impacted). The goal is to have reasonable servers in the
> pool; not to provide benchmarks for the server operators.
Yes, I know, I just look for fun. I could not explain this max. score drop
and it happened more than once in the past months.
Kind regards,
Arnold &:-)
_______________________________________________
timekeepers mailing list
[email protected]
https://fortytwo.ch/mailman/cgi-bin/listinfo/timekeepers
