[ntp:questions] ATOM falseticker with flag3 enabled

David Lord snews at lordynet.org
Sat Mar 31 21:54:36 UTC 2012


A C wrote:
> On 3/30/2012 03:30, David Lord wrote:
>> A C wrote:
>>> On 3/29/2012 03:29, David Lord wrote:
>>>> Dave Hart wrote:
>>>>> On Thu, Mar 29, 2012 at 02:44, A C <agcarver+ntp at acarver.net> wrote:
>>>>>> Without flag3 it appears that ntpd does the heavy lifting but is
>>>>>> unable to
>>>>>> use the PPS unless a prefer peer is set. With flag3 enabled, PPS is
>>>>>> running
>>>>>> without a prefer peer.
>>>>>
>>>>> Being unable to use the PPS unless either a prefer peer is set or
>>>>> another knob is adjusted (see what I quoted earlier) is operating by
>>>>> design. The change in operation with flag3 1 seems likely to me to be
>>>>> an ntpd bug, which is why I'd like to see a bug report filed. Unless
>>>>> I'm missing some design intention, it seems to me PPS should not
>>>>> become reachable with no prefer peer regardless of flag3.
>>>>
>>>>
>>>>
>>>> On NetBSD-5, i386, I had the PPS marked as falseticker as
>>>> from March 2, 2012 when I updated from 4.2.6p3 to 4.2.6p5.
>>>> When I went back to 4.2.6p3 the 'o' was back. I hadn't spare
>>>> time to investigate and changed config to use only type 20
>>>> nmea driver with 4.2.6p5.
>>>
>>> Interesting, so that's the same OS on two platforms with similar
>>> problems. Did you happen to also observe your minimum jitter times in
>>> the upgrade? When I run 4.2.6 my minimum jitter is reported as
>>> 0.061ms. But in 4.2.7 the minimum jitter is 0.122ms, exactly double
>>> the 4.2.6 value.
>>>
>>> I'm still tinkering with my configuration to see if I can get ATOM to
>>> stop being a false ticker but no luck yet even though it's still
>>> disciplining the clock.
>>
>> Jitter depends on time of day, weather etc. The variations
>> doesn't seem to have been altered by the update from 4.2.6p3
>> to p5.
>>
>> I didn't tinker as something is obviously broken with
>> type 22 driver in p5.
> 
> I understand that the jitter will change but I'm talking about the 
> minimum possible jitter that ntpd can report for a given system.  Using 
> 4.2.6, the minimum possible jitter reported (such as what is shown in 
> the ntpq billboard when ntpd first starts up) was 0.061 on my NetBSD 
> system.  However, with 4.2.7p259 (and some of the earlier versions as 
> well) the minimum possible jitter exactly doubled to 0.122 on the same 
> system.
> 
> It would appear that the method used to compute the minimum possible 
> jitter changed between the two versions.

I'm not sure which value you are referring to. From the
"ntpq -p" billboard over a day, I see values of jitter
of the GPS as low as 0.002

	jitter  No of
          0.002    156
          0.003     50
          0.004     22
          0.005      4
          0.006      4
          0.007      1
          0.012      1
          0.014      1
          0.025      1

"ntpq -c rv" for that system reports "precision=-19". ie 2^-19,
but another couple of systems without PPS source are reporting
"precision=-20" and I think that these two systems have done
better than that when temperatures have not been as varied.


David



More information about the questions mailing list