> If you dont have the chance to update your clock via ntp, then the
> RTC is *much more* accurate than the internal, interrupt driven timer-clock.
> After one or two weeks uptime, the interrupt clock had a drift of two hours,
> reported a friend of mine. We fixed this by "clock -r" every hour.
> I hope the kernel does *not* update the RTC itself , now does it ?
You've tried using adjtime to correct the problem? That is to say, are you
saying the speed of the clock varies, or the speed is unsuitable without
modification?
lilo