Hi,
First thing is to see if you're on a virtual server...
I'm not, it's a basic web server, no optimization for ntp I'm sure. Linux, based
on RedHat 7.2. Kernel is 2.4.31
Make sure your clock is actually adjusting (check for clock adjust /
reset messages in your logs)
I think it does. From the log:
Feb 6 18:17:16 ns30747 kernel: grsec: From 82.66.216.183: time set by
/usr/sbin/ntpd
Have you selected all these servers to track, or are you automatically
building associations? If you're building associations back, that may
explain the minpoll issue if you didnt set it at 4 in your config file.
I selected these, after looking at the delay:
server ntp0.nl.uu.net
server ntp1.fau.de
server ntp-p1.obspm.fr
server chronos.cru.fr
Measure once, cut twice... he is auto associating.
OK. How can I prevent this?
From sjm:
> Try deleting the file and restarting ntpd. Make sure ntp has permission
> to write to the file.
I did. IIRC, the original file was created by ntpd.
Here is my ntp.conf:
(I changed it somewhat after your comments, added minpoll, removed
broadcastdelay, removed server ntp-sop.inria.fr)
#
# stratum 1 servers
server ntp0.nl.uu.net # pas de notification [EMAIL PROTECTED]
server ntp1.fau.de
#server ntp-sop.inria.fr # notifier [EMAIL PROTECTED]
server ntp-p1.obspm.fr # notifier [EMAIL PROTECTED] ou
[EMAIL PROTECTED]
server chronos.cru.fr # notifier [EMAIL PROTECTED]
#
# minpoll
minpoll 6
#
# Drift file
driftfile /var/lib/ntp/ntp.drift
#
# Log
logfile /var/log/ntpd.log
#
# Statistics
statsdir /var/log/ntpstats/
statistics loopstats peerstats clockstats
filegen loopstats file loopstats type day enable
filegen peerstats file peerstats type day enable
filegen clockstats file clockstats type day enable
#
# Authentication
authenticate no
Just restarted ntpd, so it will take some time to stabilize.
Thanks for your help!
Christian
_______________________________________________
timekeepers mailing list
[email protected]
https://fortytwo.ch/mailman/cgi-bin/listinfo/timekeepers
