[ntp:questions] Re: GPS Truetime NTP server and offset/jitter peak

Martin Burnicki martin.burnicki at meinberg.de
Thu Oct 20 08:08:31 UTC 2005


Ralf Gross wrote:
> John Pettitt wrote:
>> Ralf Gross wrote:
>>> Martin Burnicki:
>>> 
>>> 
>>>>Let's see the capture output first. Then we'll see whether we must have
>>>>a look at the driver.
>>> 
>>> 
>>> Here's the result.
>>> 
>>> 292:23:59:59
>>> 292:23:59:60
>>> 293:00:00:00
>> 
>> 
>> !!!!!!!!!!!!!!!!!   59:60 !!!!!!!
> 
> DOH! I overlooked that :(
> 
>>> 293:00:00:01
>> 
>>> I only noticed a gap between second 21 and 23.
>>> 
>> I think the 59:60 is more the cause of the problem!  That should only
>> happen when there is a leap second.
> 
> Ok, does this mean the NTS-100-GPS FW is buggy?

Yep. 

> Should I ignore the peak, 
> because it will hopefully disappear after 12/31/2005?

If you don't get a firmware upgrade, and you can live with it, you can just
wait. 

Probably the receiver will work correctly again at the beginning of January.
Currently the GPS satellites are sending 13 and 14 for UTC offset from UTC
time before and after the leapsecond. In January, shortly after the leap
second, they will probably start to send 14 for both values and your
problems will disappear. 

You can find more details on GPS UTC/leap second parameters in two of my
earlier articles which are archived in the NTP questions mailing list
archives:
http://lists.ntp.isc.org/pipermail/questions/2005-September/006543.html
http://lists.ntp.isc.org/pipermail/questions/2005-September/006528.html

Martin
-- 
Martin Burnicki

Meinberg Funkuhren
Bad Pyrmont
Germany




More information about the questions mailing list