[ntp:questions] Regular 16ms jump of NTP on Windows XP SP2

Bert Gøtterup Petersen BUP at bang-olufsen.dk
Wed Sep 6 09:17:33 UTC 2006

I have a test setup with 12 IBM 3GHz dual core PC running either WinXp SP2 
or SuSE Linux 9.3.
I use an 'old' 933MHz Compaq PC running SuSE Linux 9.3 as NTP server 
(local clock), which otherwise does nothing.
All other PCs use this node as unicast server, with a huffpuff filter 
setting of 7200 (e.g. 2h).
All machines are connected through a professional switch from HP.

Under Windows I use NTP 4.2.0a  from Meinberg.

When running Linux all machines are synchronized within 100-200us, with an 
occasional spike of 1-2ms when a system is loaded.
This is great!

When running Windows I see a spike of about 16ms in the loopstats-files on 
all the Windows nodes. 
The peculiar thing is that the spike occurs every 24 hours and 15 minutes 
per machine, but not at the same time on different machines.

I have compiled NTP 4.2.2p for Windows, but it also shows the same spikes.

Any ideas, what can cause this?

More information about the questions mailing list