Hi guys

I have configured ntp Server on RHEl5 . The ntp server getting synchronization with live Ntp servers. But the problem is some times im getting the error of "time correction of 3220 seconds exceeds sanity limit (1000); set clock manually to the correct UTC time." and the service has been stopped. If im doing manual start to ntp service then it's working fine for next few days. What could be a reason? why it's happen?
Pls any one can suggest?

Does this happen to be a VMware server? Does the service ever start?

There is a maximum limit that time can be off from a 'real' time source, beyond that, NTP will not start. You can (as root), do an "ntpdate <ip address of a good time server>", and then try to start NTP. That should get you close enough for the service to start.

If it's a VMware server, there is a known bug with RHEL and VMware, causing the time to drift. RedHat has a patch available for it, along with download instructions.

Hi Acid,
Thank you so much for your advice,

I added on the synchronization sources & remove fudge.

#allow synchronization sources from the internet servers
server 2.sg.pool.ntp.org
server 0.asia.pool.ntp.org
server 2.asia.pool.ntp.org

REMOVE
server 127.127.1.0 #local system clock
fudge 127.127.1.0 stratum 10

and it's working.
#allow synchronization sources from the internet servers
server 2.sg.pool.ntp.org
server 0.asia.pool.ntp.org
server 2.asia.pool.ntp.org

#restict the type of access for each internet server
restrict 2.sg.pool.ntp.org mask 255.255.255.255 nomodify notrap noquery
restrict 0.asia.pool.ntp.org mask 255.255.255.255 nomodify notrap noquery
restrict 2.asia.pool.ntp.org mask 255.255.255.255 nomodify notrap noquery

#allow our internal server to synch to this server
restrict 192.168.1.0 mask 255.255.255.0 nomodify notrap noquery notrust

driftfile /etc/ntp/drift
logfile /var/log/ntp.log

--------------------------------------------------------------------------------
[root@app3 ~]# ntpq -p
remote refid st t when poll reach delay offset jitter
==============================================================================
starlite2.ispwo 204.152.184.72 2 u 10 64 3 11.924 3.807 1.980
filter.datautam 195.43.74.3 2 u 36 64 1 93.940 51.346 0.001
158.108.2.100 158.108.7.157 2 u 42 64 1 226.146 99.108 0.001

I do have another issue here, what if the internet time server is down. i won't be able to synch to them and i need to fall back on something. Am i right to say the entry below is something similar for falling back on the local system?
what is stratum mean?

server 127.127.1.0 # local clock
fudge 127.127.1.0 stratum 12

No one named Acid has posted on this thread
The 'fudge' means that your server will drop to stratum 12, if it can't find the time from further upstream
Stratum means level of accuracy. 0 is best, 16 is worst. And do not reopen old threads..this thread is TWO YEARS OLD, and especially don't reopen old threads, then hijack them with your own question. Open your own thread.

Hi TBOne,

Thank you so much for the information,
So Sorry to dig out a old thread.