[ntp:questions] Re: Are these Windows XP clients beyond hope?

David J Taylor david-taylor at blueyonder.co.not-this-bit.nor-this-part.uk.invalid
Mon Sep 19 11:54:23 UTC 2005


Neil Trotter wrote:
[]
> In fact, you already did help, David.  I visited your site yesterday,
> and still have a window open on it :-)  I downloaded NTPMonitor to
> give me an overview, and am planning to read more today.

OK, I use MRTG to plot my own data, based on the output from an NTPQ 
command.

[]
> They are "standard" PCs, specs follow.  I believe that either of these
> boards may be used for overclocking, but I assembled the PCs myself,
> and prefer reliability over speed, so I know they are running at the
> design frequency.  Is it possible that e.g. a clock-multiplier
> setting might cause problems?  I had thought this didn't alter the
> speed of the clock itself, but am not an overclocking hobbyist, so
> don't know too much about such things.

Overclocking shouldn't be an issue, I think.  I was wondering more about 
settings like powering down (or lowering the CPU frequency) when not in 
use.

[]
> The A7N8X has an identical twin "sister" (component-wise) on the
> network, which keeps good time.  That is to say, all components were
> purchased at the same time from the same vendor, so I *assume* they
> are identical, though there is obviously *some* variable factor in
> the mix, be it hardware or configuration.

My guess is software rather than hardware.

>> - if you do a ntpq -p on those systems, do they appear to be in sync?
>
> No, they don't; this is how I was monitoring them.

Ah, so is the problem that these PCs are not seeing their NTP servers 
properly.  Perhaps you could elucidate on this point?  How are the PCs 
"not syncing"?  If there was a batch of bad crystals such that the clock, 
when free-running, exceeded the 500ppm error which NTP allows, that would 
be an issue.

[]

Cheers,
David 





More information about the questions mailing list