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

Martin Burnicki martin.burnicki at meinberg.de
Tue Oct 18 09:58:03 UTC 2005


Hi,

Ralf Gross wrote:
> Hi,
> 
> I've a problem with a Truetime NTS-100-GPS device connected by RS232 to a
> CentOS 4 Linux server (ntp-4.2.0.a.20040617-4).
> 
> Every night at 2am (CEST) offset an jitter suddenly shows very high values
> and as a result ntpd loses sync to the GPS stratum 0 device for about 30
> seconds.

2am CEST is UTC midnight ...

[...] 
> 02:00:08
>       remote       refid   st t when poll reach   delay   offset  jitter
>
==============================================================================
> xTRUETIME(0)     .TRUE.    0 l    1   16  377    0.000  -999.57   999.525
> ...                                                     ^^^^^^^
> ---------------------------------- peak ------------------/

This is about exactly -1 second. Did you hear about the upcoming leap second
which is scheduled for UTC midnight after Dec 31, 2005?

The leap second is currently already announced in the GPS satellite data
stream, and if the GPS receiver doesn't handle the announcement correctly
it might introduce the leap second every UTC midnight. 

We've read here about at least one other type of GPS receiver which was
going to insert the leap second at the end of September instead of the end
of December.

Do you have the possibility to log the GPS serial output across UTC midnight
in order to verify that?


Martin
-- 
Martin Burnicki

Meinberg Funkuhren
Bad Pyrmont
Germany




More information about the questions mailing list