Dear David,
> BTW: I would be most interested to see your code for getting ...
I will publish this piece of code.
It's a bash/gawk/gnuplot script
// hans
--
Am 05.04.2016 08:43, schrieb David J Taylor:
dear all,
i run 2 NTP server on 2 identical systems. both are
banana pi's M1 with GPS modules from M0UPU with a small GPS
mouse ( GPS antenna ).
this configuration i run on 2 different locations - let
say A ( a rurally area ) and B ( in an urban area ) - since
about last June successfully. typically there are between 6 and
7 satellites visible over 24 hour's in average which gives a
stable time information for NTP.
and sometimes even much more are available.
this days i recognized a dramatically change at march 24th
around lunch time on location A. currently less than 2 satellites
are visible in average for 24 hours and over long periods even
no one. in summary 11 hours at 03/25
i will attach 2 png's to see the difference between date
before and after 03/24 of usable satellites.
( this will be hopefully accepted by the mailing list )
[]
have you ever seen such a strange behaviour ?
i can only guess there is a source of interference.
but how to find it ?
any feedback or any ideas are welcome.
kind regards
hans
===========================
Hans,
You have done all the tests I would have suggested, and it points to
interference at one location. Suggest someone is operating a GPS jammer
illegally. You could try locating the source with a suitable receivers
(look for broadband noise around 1575.42 MHz) but be careful if there
may be
criminal activity involved.
One other possibility - vegetation growth during spring and summer can
reduce the GPS signals. Be sure to test with the antenna outdoors in
case
something has been deposited on the roof.
BTW: I would be most interested to see your code for getting the
number of
satellites, as I would like to add that parameter to my MRTG graphs.
Cheers,
David