GPS Clock problem sept 3 2016, around 20:14 EST

[bourque...@gmail.com]

Were they any problem reported yesterday night with GPS clock source? Our 2 very old (+15 years) Lantronix GPS clock (on 2 building, 15 km apart) drift/jump simultaneously at 20h14

By the time I was advised, they were returning time with an 800 sec offset around 21:45 , to finally resync about 20 min later. 2 Internet NTP sources keep the network running without problem.

Anybody saw something or my GPS clock are too old and saw something they wrongly interpret???

[David Taylor]

No jumps here with multiple different GPS/PPS devices, times in UTC:

http://www.satsignal.eu/mrtg/performance_ntp.php

However, all these PPS systems also have Internet servers as well as
some using the NMEA output from the GPS. I don't currently have any
completely stand-alone devices (except for the LeoNTP server).

--
Cheers,
David
Web: http://www.satsignal.eu

[bourque...@gmail.com]

I though I saw one of the clock with a jump at the same time as my problem but I wrongly calculed the time offset for GMT

[bourque...@gmail.com]

Le dimanche 4 septembre 2016 12:33:05 UTC-4, David Taylor a écrit :

And, I just realized you are the David Taylor that make the NTPMonitor tool I am using!!! Thank you for the tool, it help me a lot yesterday night

[David Taylor]

On 04/09/2016 17:51, bourque...@gmail.com wrote:
[]
> Well in fact, one of the GPS base clock show a hit at the same time I was affected:
> PC Kiruna
> Intel i7-4770K 16 GB
> Windows-10/64 Pro
> Interactive
> Stratum-1 GPS/PPS kernel-mode sync
> Sure GPS, antenna indoors

Noted, and on the wrong offset, but in fact PC Kiruna was rebooted then:

http://www.satsignal.eu/mrtg/performance_kiruna.php

Glad you're enjoying the software!

[Martin Burnicki]

Do you know which type of GPS receiver is assembled in your clock? A
problem with a Trimble Acutime 2000 has been reported on the NTP pool
mailing list:
https://lists.ntp.org/pipermail/pool/2016-September/007944.html

Martin
--
Martin Burnicki

Meinberg Funkuhren
Bad Pyrmont
Germany

[bourque...@gmail.com]

Yes this is the Trimble with an RS-422 interface bought more then 15 years ago
The Lantronix NTS doesn't give access to any data so I don't know what happen exactly except the 2 of them were returning very bad data and were considered insane, invalid by our Cisco 6509 and the time frame is the same then the one they are reporting