[ntp:questions] Stratum goes back to 16 all the time
Richard B. Gilbert
rgilbert88 at comcast.net
Sun Jan 14 16:41:21 UTC 2007
> Harlan Stenn wrote:
>>If your stratum is at 16 it means your machine is not syncing with anything.
>>Have you read and followed the instructions at http://ntp.isc.org/Support
>>(especially the chapters on Configuring NTP, Starting NTP, and
> I changed some configuration options, like adding iburst etc that was
> suggested, so it's not stratum 16 as often as before, but it still goes
> back to 16 every now and then.
> It seems like my server is unable to answer to queries sometimes, maybe
> because of packetloss.
> I got an email off this list where someone said that NTP was designed
> when networks was usually not very stable, and packetloss was common.
> But still, my server doesn't seem to handle this very well.
> This is my peers list now:
> ntpq> peers
> remote refid st t when poll reach delay offset
> -nissan.ifm.liu. .PPS. 1 u 12 64 3 28.275 -259.30
> +timmy.lysator.l .GPS. 1 u 33 64 3 29.503 -234.20
> +timehost.lysato .GPS. 1 u 24 64 3 30.316 -242.76
> *Time1.Stupi.SE .PPS. 1 u 45 64 3 24.547 -228.69
> +ntp1.sth.netnod .PPS. 1 u 17 64 3 30.314 -244.02
> -ntp2.gbg.netnod .PPS. 1 u 5 64 7 30.375 -218.94
It appears, from the ntpq banner, that ntpd has been running for, at
most, three or four minutes! Ntpd has selected a synchronization source
but has not yet corrected a very substantial offset.
Let that run for at least thirty minutes and show us the output of ntpq
-peers again. If that display was taken after ntpd had been running for
more that three or four minutes, there is something wrong with your
More information about the questions