[ntp:questions] Re: ntp sanity limit kills ntp daily

Bill Davidsen davidsen at deathstar.prodigy.com
Mon May 23 16:35:31 UTC 2005

Michael Ward wrote:
> Hello All,
> I've configured my FC3 system to run ntp successfully at boot time and 
> ntp seems to run successfully for about a day.  However, every night the 
> clock looses some 1200 seconds while the system is busy making backups. 
>  When ntp figures this out, it refuses to update the clock because this 
> is classified as insane, and terminates.
> I happen to agree that the clock loosing about 1200 seconds (for any 
> reason) seems insane.  However, given that this sort of behavior seems 
> common enough to be found via google (search for: ntp exceeds sanity 
> limit), I'm hopeful that there is a way to coax ntp (or some other 
> automated program) to trudge on and do the deed anyway.
> Default rules can make is seem like a crazy world, can't they?
> I'm running a self compiled linux kernel 2.6.9 with FC3 linux as 
> installed from Negus' "Red Hat Fedora Linux 3".  Anyoune out there know 
> how to make this sort of thing work in a rational manner?

Look for a device not using DMA, or not allowing interrupts while i/o is 
in progress. If you back up to CD/DVD in pio mode you risk loss of time 
and data integrety (have you checked you backups?) doing that. Check 
/proc/irq and see if you have a lot of shared irq's, or have you diddles 
acpi (or should you?).

Just some of the things you can think about, may well be "none of the 

bill davidsen
   SBC/Prodigy Yorktown Heights NY data center

More information about the questions mailing list