[ntp:questions] Accuracy of GPS device

unruh unruh at wormhole.physics.ubc.ca
Fri Sep 2 06:56:28 UTC 2011


On 2011-09-02, Miguel Gon?alves <mail at miguelgoncalves.com> wrote:
> Comments bellow...
>
> On 2 September 2011 00:47, unruh <unruh at wormhole.physics.ubc.ca> wrote:
>
>> On 2011-09-01, Miguel Gon?alves <mail at miguelgoncalves.com> wrote:
>> > Hi!
>> >
>> > Thanks for your reply. My comments bellow.
>> >
>> > On 1 September 2011 18:24, unruh <unruh at wormhole.physics.ubc.ca> wrote:
>> >
>> >> On 2011-09-01, Miguel Gon?alves <mail at miguelgoncalves.com> wrote:
>> >> > Hi all!
>> >> >
>> >> > I have two internal FreeBSD with GPS receivers attached (Garmin 18
>> LVC:
>> >> > 10.0.2.10 / Sure Evaluation Board:10.0.2.9). Both machines are on the
>> >> same
>> >> > LAN segment (VLAN).
>> >> >
>> >> > For redundancy, I've configured a Cisco switch as a stratum 2 server.
>> >> Here's
>> >> > the relevant information:
>> >> >
>> >> > $ ntpq -pcrv 10.0.2.254
>> >> >      remote           refid      st t when poll reach   delay   offset
>> >> >  jitter
>> >>
>> >>
>> >==============================================================================
>> >> > +ntp0.as34288.ne .PPS.            1 u  814 1024  377   72.750
>> -1.084
>> >> > 0.780
>> >> > +canon.inria.fr  .GPSi.           1 u  399 1024  377   55.110
>>  0.218
>> >> > 0.400
>> >>
>> >> What are those machines? You have names rather than IP addresses.
>> >> Are they your pps machines?
>> >
>> >
>> > No. This is a stratum 2 server and it gets the time from stratum 1
>> servers
>> > thus the names and not IP addresses.
>>
>> What I am asking is what the mapping is between these names and the
>> numbers you have later. I assume that some of those names are the same
>> machine as the IP addresses you list below but We do not have that
>> infomation.
>>
>
> Which numbers?
>
> 10.0.2.254 is a stratum 2 server that gets time information from outside
> servers (I've included the information). canon.inria.fr is a server in
> France that has a GPS clock receiver AFAIK.
>
> 10.0.2.254 is just a Cisco switch that I set up internally for comparison to
> the the stratum 1 servers (10.0.2.9 and 10.0.2.10).
>
> 10.0.2.2 is another server (Linux, CentOS) that gets time from 10.0.2.9,
> 10.0.2.10 and 10.0.2.254. I've written all this before. I thought I was
> clear enough.

Nope.

It is completely unclear to me what your question is. Your 10.0.2.254 is
an outside switch.

>
>
>> > $ ntpq -p 10.0.99.99
>> >      remote           refid      st t when poll reach   delay   offset
>> >  jitter
>>
>> >==============================================================================
>> > *10.0.2.10       .GPS.            1 u   21  256  377    0.173    0.196
>> > 0.008
>> > +10.0.2.9        .GPS.            1 u   93  256  377    0.175    0.191
>> > 0.014
>> > +10.0.2.254      81.94.123.16     2 u  149  256  377    0.583   -6.884
>> > 0.152

This tells me that your two GPS receivers are consistant with each
other, but I have no idea why the offset is larger than the delay, and
why the offsets are so large. On a lan, the offsets should be a factor
of 20 or so less than what you are getting here. 
That the external router is 7 ms out just tells me that it is really
poorly synced with the outside world. 


>> >
>> > This is a FreeBSD embedded PBX machine running Asterisk. The machine is
>> > mostly idle. What kind of offsets should I get with local machines?
>>
>> in the 10s of usec range max. Certainly less than the delay.
>>
>
> tens of usec is good... Anyone here which can explain why NTP isn't getting
> that?

How could we? 
Maybe you are running a virtual BSD machine, and thus the clocks are
wonkey. Maybe you have lousy hardware. Who knows. 


>
>
>>  > Assuming ntp04, ntp05 and ntp06 are on the same LAN I see offsets higher
>> > than 100 us. Is it possible to decrease these numbers?
>>
>> Sure. all my systems have offsets in the 10us range--  on the same lan
>> as my time server.
>> Mind you I do use chrony, not ntpd but even ntpd should be in a few 10s
>> of usec.
>
>
> Can ntpd really get there? I'll try to query some good public servers to see
> what others are getting...

Sure it can. It can get better than 30us. But why you are not getting it
is impossible for us to say.  

>
>> All my machines are Linux machines. Linux is fine for timekeeping.
>>
>
> OK. No criticism intended. I'm just a BSD guy.




More information about the questions mailing list