Hi!
I am been participating in the NTP Pool Project for a week and I like
contributing to the community.
Today, unfortunately my firewall blocked all traffic between 00:30 UTC
and 09:15 UTC. I was making some changes to be able to add a ADSL
connection whose traffic will be redirected to a local stratum 1
server in our DMZ network.
I already have an internal server responding to a public IP address.
When I add the public IP address from the ADSL what is the best
approach: pointing the IP address to the current stratum 1 or should I
add another machine? The current machine has a lot of CPU cycles to
spare and keeps time within 10 us of UTC:
$ ntpq -p 172.16.0.1
remote refid st t when poll reach delay offset jitter
==============================================================================
*GPS_NMEA(0) .GPSg. 0 l 2 16 377 0.000 7.612 4.578
oPPS(0) .PPSg. 0 l 1 16 377 0.000 -0.007 0.004
+ntp-p1.obspm.fr .TS-3. 1 u 44 64 377 44.847 1.048 0.896
+ptbtime1.ptb.de .PTB. 1 u 5 64 377 65.351 0.756 1.689
+ntp1.oma.be .PPS. 1 u 4 64 377 50.603 2.824 0.730
+canon.inria.fr .GPSi. 1 u 34 64 377 45.076 1.423 1.013
+ntp1.nl.uu.net .PPS. 1 u 3 64 377 54.489 1.790 1.987
+ntp1.as34288.ne .PPS. 1 u 44 64 377 61.478 1.592 1.384
+smtp.certum.pl .GPS. 1 u 39 64 377 84.244 -1.174 0.963
Regarding the current 100 ms target accuracy of the pool I was
surprised today when I requested the time from 0.pt.pool.ntp.org and
compared it to my machine that as a Motorola Oncore UT+ GPS receiver:
oncore# ntpdate -q 0.pt.pool.ntp.org; ntpq -p
server 81.92.212.46, stratum 3, offset -0.034996, delay 0.03427
server 194.117.9.136, stratum 2, offset -0.035880, delay 0.03557
server 212.113.164.13, stratum 4, offset -0.000382, delay 0.03668
20 Oct 18:52:23 ntpdate[18244]: adjust time server 194.117.9.136
offset -0.035880 sec
remote refid st t when poll reach delay offset jitter
==============================================================================
oGPS_ONCORE(0) .GPS. 0 l 6 16 377 0.000 0.007 0.004
+pal-213-228-178 .PPSg. 1 u 12 16 377 18.840 -0.641 0.745
dial-b1-212-203 .PPSg. 1 u 161m 16 0 23.096 0.016 0.000
-ntp02.oal.ul.pt 194.117.9.137 2 u 48 64 377 10.069 -3.650 0.273
-ntp04.oal.ul.pt 194.117.9.138 2 u 66 64 377 10.221 -38.436 1.656
+Router7.Lisboa. 193.136.250.246 2 u 61 64 377 8.575 0.554 0.277
+Router15.Porto. 193.136.250.246 2 u 1 64 377 12.954 0.475 0.247
Apparently I have a good clock (GPS) but the pool is saying my clock
is 35 ms off.
A few minutes later I tried to query the French and Portuguese servers
and I got this
oncore# ntpdate -q 0.fr.pool.ntp.org ; ntpdate -q 0.pt.pool.ntp.org
server 91.121.162.168, stratum 2, offset 0.003431, delay 0.08351
server 193.55.167.2, stratum 2, offset 0.000632, delay 0.07948
server 213.186.39.202, stratum 3, offset 0.008381, delay 0.08681
20 Oct 19:40:16 ntpdate[18397]: adjust time server 193.55.167.2 offset
0.000632 sec
server 194.117.9.136, stratum 2, offset -0.006412, delay 0.03580
server 84.90.89.37, stratum 3, offset 0.006486, delay 0.05521
server 212.113.164.13, stratum 3, offset -0.000483, delay 0.03661
20 Oct 19:40:17 ntpdate[18398]: adjust time server 194.117.9.136
offset -0.006412 sec
632 us from France and 6412 us from local servers? Fortunately I have
GPS clocks :-) Yes, I know that this project's target is not the
ultimate accuracy but a 10 times bigger error is difficult for me to
understand.
Please don't get me wrong as this is just my humble opinion but
perhaps 100 ms whas a good target when the project started. Why don't
we get this down to 25 ms? Since today everyone is requesting time
from servers close by I believe 25 ms (or even 10) would be a nice
target.
Cheers,
Miguel
_______________________________________________
pool mailing list
[email protected]
http://lists.ntp.org/listinfo/pool
