[ntp:questions] NTPD can take 10 hours to achieve stability

unruh unruh at wormhole.physics.ubc.ca
Wed Apr 20 01:20:29 UTC 2011


On 2011-04-19, David Lord <snews at lordynet.org> wrote:
> unruh wrote:
>> On 2011-04-19, David Lord <snews at lordynet.org> wrote:
>>> David J Taylor wrote:
>>>> "unruh" <unruh at wormhole.physics.ubc.ca> wrote in message 
>>>> news:slrniqpa40.ppu.unruh at wormhole.physics.ubc.ca...
>>>> []
>>>>> Well, I think someone other than the current maintainers will have to
>>>>> port it to windows. Since windows timekeeping is not the worlds best
>>>>> anyway, it is probably true that the extra accuracy of chrony is
>>>>> unnecessary. It does have a command line option "like ntpq -p"
>>>>> provided by chronyc (depending on what you mean by "like").
>>>>> What MRTG is I do not know.
>>>>>
>>>>> If you are happy with ntpd, by allmeans stay with it.
>>>> A pity that chrony will not be offered for Windows, at least for tests 
>>>> to see whether it lives up to its claims.  There are times when a more 
>>>> rapid convergence would be welcome, such as the reboot of PC Molde 
>>>> around 13:30 yesterday:
>>>>
>>>>  http://www.satsignal.eu/mrtg/molde_ntp-b.html
>>>>
>>>> MRTG is a standard logging tool for network I/O which uses SNMP to 
>>>> produce the graphs I have quoted here many times for network throughput 
>>>> and timekeeping:
>>>>
>>>>  http://www.satsignal.eu/mrtg/performance_network.php
>>>>  http://www.satsignal.eu/mrtg/performance_ntp.php
>>>>
>>>> I've written how to extend MRTG to monitor NTP timekeeping, and various 
>>>> other parameters such as disk space and temperature here:
>>>>
>>>>  http://www.satsignal.eu/ntp/NTPandMRTG.html
>>>>  http://www.satsignal.eu/mrtg/performance_howto.php
>>>>
>>>>  From chronyc I would need to be able to use a simple Perl script to 
>>>> extract the numbers to be plotted - such as the Offset in the graphs 
>>>> above.  An easy job if the format is standardised and machine readable.
>>> I based my script on yours for ntp but changed from using a
>>> linear axis to log. I first used chrony on NetBSD when on
>>> dialup with demon and later for a while when on broadband.
>>> When I first compared chrony with ntpd there was no contest
>>> but more recent experiments with chrony had periods of
>>> severe instability much worse than ntpd.
>> 
>> More recent means what? What version of chrony? (there was a bug found
>> in the past two weeks that was introduces a few months ago which did
>> result in instability) 
>
> More recent than  November 2009.
> Probably between Dec 2009 and Jan 2010 and p4x2400c.
>
> chrony 1.23

OK well before that change. 

>
>
>>> P4X2666 with chrony
>>> <http://www.lordynet.org.uk/mrtg/stats/>
>> 
>> Not at all sure what I am supposed to see. I have no idea what the graph
>> axes represent?  What is 1.1k 
>
> X-axis is in hours
> Y-axis is offset in us so 1.1k = 1100us

WOW. What kind of network are you attached to? Even on an ADSL link
through the phone company, I
was getting in the tens of usec (not ms) as the offsets of chrony.
(checked by a gps receiver attached to the local computer).

And what is it about the graph that makes you believe you are seeing
instability. 


>
> Next experiment is to couple up the GPS again but if you
> click on the ME6000 in the current stats the yearly graph
> for Sep/Oct/Nov was when I last had the gps connected.
>
>
> David
>
>> Note that you can see my offsets with chrony on my web page. The network
>> problems really do cause difficulty, but the offsets in general are very
>> small -- 10usec over the network. 
>> 
>> 
>>> Current stats all with ntpd.
>>> <http://www.lordynet.org.uk/mrtg/stats/ntp/>
>>>
>>>
>>> David
>>>




More information about the questions mailing list