NTP problem for virtual RHEL 4 server on VmWare

[Kenneth Holter]

Hei.


One of our RHEL 4 servers running on VmWare has a quite serious NTP problem.
I know that NTP can be an issue when running red hat boxes on VmWare, so as
a fix I put this small script in a file in /etc/cron.hourly:


[root at server cron.hourly]# cat ntpdate
#!/bin/sh
/etc/init.d/ntpd stop
ntpdate 1.2.3.4 >> /tmp/time_adjust.log
/etc/init.d/ntp


After investigating the "/tmp/time_adjust.log" file, I was quite surprised
by the amount of drift found on one particular server. Consider this extract
from the file:

6 Nov 20:00:01 ntpdate[19373]: step time server 1.2.3.4 offset -60.504153
sec
 6 Nov 20:00:52 ntpdate[19666]: step time server 1.2.3.4 offset -8.735440
sec
 6 Nov 20:01:00 ntpdate[19689]: step time server 1.2.3.4 offset -1.635632
sec
 6 Nov 20:54:06 ntpdate[24198]: step time server 1.2.3.4 offset -415.894712
sec
 6 Nov 21:01:01 ntpdate[24920]: adjust time server 1.2.3.4 offset 0.136833
sec
 6 Nov 22:01:02 ntpdate[29943]: adjust time server 1.2.3.4 offset -0.114253
sec
 6 Nov 23:01:01 ntpdate[2519]: adjust time server 1.2.3.4 offset -0.036345
sec
 7 Nov 00:01:00 ntpdate[7577]: step time server 1.2.3.4 offset -1.064935 sec
 7 Nov 01:00:57 ntpdate[12697]: step time server 1.2.3.4 offset -3.922577
sec
 7 Nov 02:00:21 ntpdate[17733]: step time server 1.2.3.4 offset -40.421825
sec
 7 Nov 02:01:00 ntpdate[17777]: step time server 1.2.3.4 offset -1.123175
sec
 7 Nov 02:57:23 ntpdate[22542]: step time server 1.2.3.4 offset -218.649820
sec
 7 Nov 03:00:36 ntpdate[22900]: step time server 1.2.3.4 offset -25.284528
sec
 7 Nov 03:00:58 ntpdate[22940]: step time server 1.2.3.4 offset -3.104130
sec
 7 Nov 03:52:32 ntpdate[27430]: step time server 1.2.3.4 offset -509.363952
sec
 7 Nov 03:59:50 ntpdate[27943]: step time server 1.2.3.4 offset -71.430354
sec
 7 Nov 04:00:52 ntpdate[28236]: step time server 1.2.3.4 offset -9.344907
sec
 7 Nov 04:01:00 ntpdate[28259]: step time server 1.2.3.4 offset -1.237651
sec
 7 Nov 05:01:01 ntpdate[1363]: adjust time server 1.2.3.4 offset 0.390149
sec
 7 Nov 06:01:01 ntpdate[6419]: adjust time server 1.2.3.4 offset -0.185112
sec
 7 Nov 07:01:02 ntpdate[11493]: adjust time server 1.2.3.4 offset -0.228884
sec
 7 Nov 08:00:59 ntpdate[16579]: step time server 1.2.3.4 offset -2.166519
sec
 7 Nov 09:00:38 ntpdate[21522]: step time server 1.2.3.4 offset -23.169420
sec
 7 Nov 09:01:02 ntpdate[21558]: adjust time server 1.2.3.4 offset -0.492106
sec
 7 Nov 09:59:26 ntpdate[26329]: step time server 1.2.3.4 offset -95.154264
sec
 7 Nov 10:00:55 ntpdate[26639]: step time server 1.2.3.4 offset -5.997955
sec
 7 Nov 10:01:01 ntpdate[26658]: step time server 1.2.3.4 offset -0.506367
sec


Does anyone know what may be causing the RHEL box to drift as much as 500
seconds in only one hour?

Regards,
Kenneth Holter