[ntp:questions] how did ntp service set the maxallowphaseoffset
Martin Burnicki
martin.burnicki at meinberg.de
Mon Nov 25 14:36:19 UTC 2013
xiaoniao112233 at gmail.com wrote:
> hello :
> I had recently start a work about ntp service ,my friends and me use
> windows and linux to sync time in ntp.we could use w32time service to
> sync linux in ntp.
I hope the Linux machine is the NTP server, and your Windows machine is
the NTP client. Linux and other Unix-like systems are usually much
better timekeepers than Windows is.
> In windows we could I found that in the registry:
> HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\W32Time\Config\
> MaxAllowedPhaseOffset may be could set the offset time between the
> local time and the server that they will be synchronize. If out of
> range of MaxAllowedPhaseOffset in seconds ,they synchronize,otherwise
> not. But I have a question that Is there a paremeter in ntp service
> that control the offset like in registry above in windows.
As has been noted by others here, ntpd tries to slew the system time
smoothly if the estimated time difference is below a certain limit, and
just steps the system time if if the determined offset exceeds this
limit for some time. The limit is usually 128 ms, but can be changed.
Unfortunately you don't tell which Windows version and which version of
the NTP service you are using.
If you are encountering problems getting the Windows time adjusted
properly by ntpd then a a Windows bug can be the reason why.
Some Windows versions don't apply small time adjustments to the system
time at all. For example, if NTP applies an adjustment less than 16
ticks to the Windows time this is simply ignored by Windows. However,
NTP expects the adjustment to have some effect, but if there is no
effect then the next time comparison yields a much larger difference
than expected, and thus causes another adjustment which is probably
larger than necessary. As a summary this can cause large swings in the
time adjustment values.
A developer version of the NTP package contains a workaround for this
Windows bug. The report and fix are discussed here:
NTP Bug 2328 - Vista/Win7 time keeping inaccurate and erratic
https://bugs.ntp.org/show_bug.cgi?id=2328
The bug report contains a link to the Microsoft support web page
explaining the bug:
SetSystemTimeAdjustment May Lose Adjustments Less than 16
http://support.microsoft.com/kb/2537623
Even though the MS report only mentions Windows 7, the Windows Server
2008 kernel is similar to Windows 7 and has probably the same bug. So if
you want to give it a try you can download a developer version of the
NTP package which includes a workaround here:
http://support.ntp.org/people/burnicki/windows/
You should try the release version first. Just unzip the ZIP archive,
stop the NTP service, copy all extracted files over the files in your
NTP installation directory (e.g. C:\Program Files (x86)\NTP\bin\), and
restart the NTP service.
Also all newer development versions of the NTP binaries should include this.
We have found that this version has greatly improved the resulting
accuracy on Windows 7 and Windows Server 2008 installations.
Please note under Windows you should configure all upstream servers with
a line reading
server aa.bb.cc.dd iburst minpoll 6 maxpoll 6
where aa.bb.cc.dd has to be replaced with the host name or IP address of
your NTP server.
Please don't use polling intervals below 6 with the developer version
since this prevents the workaround from working correctly as discussed
in the bug report.
Also, if you don't limit the upper bounds of the polling interval by
"maxpoll 6" you may run into this bug:
NTP Bug 2341 - ntpd fails to keep up with clock drift at poll > 7
http://bugs.ntp.org/show_bug.cgi?id=2341
On the other hand, the developer version above configured with regard to
the hints above has fixed the timekeeping and significantly increase the
time accuracy in a number of installations of Windows 7 and Windows
server 2008.
Hope this helps.
Martin
--
Martin Burnicki
Meinberg Funkuhren
Bad Pyrmont
Germany
More information about the questions
mailing list