im getting errors on all of my NTP clients.
Oct 27 14:44:15 cms2 ntpd[19830]: synchronisation lost
Oct 27 15:08:25 cms2 ntpd[19830]: time reset 0.688591 s
Oct 27 15:08:25 cms2 ntpd[19830]: synchronisation lost
Oct 27 15:28:45 cms2 ntpd[19830]: time reset 0.462257 s

i checked several things:

ntpq> rv
status=0644 leap_none, sync_ntp, 4 events, event_peer/strat_chg,
version="ntpd 4.1.2@1.892 Tue Feb 24 06:32:25 EST 2004 (1)",
processor="i686", system="Linux2.4.21-32.0.1.ELsmp", leap=00, stratum=3,
precision=-17, rootdelay=20.179, rootdispersion=33.180, peer=29037,
refid=ntpServer1,
reftime=ccb03d7c.ac03ff69 Mon, Oct 27 2008 14:15:40.671, poll=6,
clock=ccb03d84.6b1044f1 Mon, Oct 27 2008 14:15:48.418, state=4,
offset=-96.471, frequency=442.496, jitter=98.288, stability=5.497

[root@cms1 root]# tcpdump -n udp port 123
tcpdump: listening on bond0
10:24:46.786062 10.23.1.30.ntp > 10.23.1.4.ntp: v4 client strat 3 poll 6 prec -15 (DF) [tos 0x10]
10:24:46.786975 10.23.1.4.ntp > 10.23.1.30.ntp: v4 server strat 2 poll 6 prec -17 (DF)
10:24:54.786359 10.23.1.30.ntp > 10.23.1.3.ntp: v4 client strat 3 poll 6 prec -15 (DF) [tos 0x10]
10:24:54.786984 10.23.1.3.ntp > 10.23.1.30.ntp: v4 server strat 2 poll 6 prec -17 (DF)

root@cms3 root]# ntpq -p
remote refid st t when poll reach delay offset jitter

LOCAL(0) LOCAL(0) 5 l 42 64 1 0.000 0.000 0.061
+ntpServer1 172.16.21.210 2 u 18 64 1 0.373 -15.214 3.043
*ntpServer2 172.16.21.210 2 u 19 64 1 0.341 -10.955 2.306

any suggestions?
cheers

from what i read its happen when there is a more then 128ms delay.
is it possible to change the thrshold of 128ms ?

The message means that your clock was corrected by more than 128ms. Your clock about 7 tenths of 1 second off in the worst case that you show. Is this really a big problem? If so, how much accuracy is actually required? I wouldn't take any action unless the clock was more than one full second off.

Since your jitter seems low, my guess is that your system's clock is running slow. Talk to the manufacturer. There may be a procedure to adjust it.

ya i know its not critical
but my main issue is to get ride of the alarm:
anyway do you happen to know how can this specific error message can be ignored by the /var/log/messages ?
Cheers