Linux ntp client not working

[herzel]

Hi All
We have in our lan amount of something like 500 servers / workstations .
Mix of Unix,Linux and windows machines.
We have Active Directory implemented and our primarey time server is windows.
We have no any problem with ntp implementation at any Unix or Windows machines.
The only problem that exist is with Linux machines.
Seems like Linux machines act as like as they want.
I do configure /etc/ntp.conf like following

server 141.226.76.3
driftfile /var/lib/ntp/drift

After that do service ntpd restart (RedHat)
Nothing does happen .
The output for the above is :

service ntpd restart
Shutting down ntpd: [ OK ]
Starting ntpd: [ OK ]
But all stay as same
Nothing change !
The /var/log/messages sayes:
"
"Mar 24 17:11:07 tlplsxa01 ntpd[14236]: Listening on interface wildcard, ::#123
Mar 24 17:11:07 tlplsxa01 ntpd[14236]: Listening on interface lo, 127.0.0.1#123
Mar 24 17:11:07 tlplsxa01 ntpd[14236]: Listening on interface eth0, 141.226.76.71#123
Mar 24 17:11:07 tlplsxa01 ntpd[14236]: kernel time sync status 0040
Mar 24 17:11:07 tlplsxa01 ntpd[14236]: frequency initialized 0.000 PPM from /var/lib/ntp/drift
Mar 24 17:17:31 tlplsxa01 ntpd[14236]: ntpd exiting on signal 15
Mar 24 17:17:31 tlplsxa01 ntpd: ntpd shutdown succeeded
Mar 24 17:17:31 tlplsxa01 ntpd[14769]: ntpd 4.2.0a@1.1190-r Mon Oct 11 09:10:20 EDT 2004 (1)
Mar 24 17:17:31 tlplsxa01 ntpd[14769]: precision = 1.000 usec
Mar 24 17:17:31 tlplsxa01 ntpd[14769]: Listening on interface wildcard, 0.0.0.0#123
Mar 24 17:17:31 tlplsxa01 ntpd[14769]: Listening on interface wildcard, ::#123
Mar 24 17:17:31 tlplsxa01 ntpd[14769]: Listening on interface lo, 127.0.0.1#123
Mar 24 17:17:31 tlplsxa01 ntpd[14769]: Listening on interface eth0, 141.226.76.71#123
Mar 24 17:17:31 tlplsxa01 ntpd[14769]: kernel time sync status 0040
Mar 24 17:17:31 tlplsxa01 ntpd: ntpd startup succeeded
Mar 24 17:17:31 tlplsxa01 ntpd[14769]: frequency initialized 0.000 PPM from /va"

I know this complain is very strange becuase it not possible ntp not working with linux but in our lan this is afact

there is nothing mentinoed in the log regared the real time server : 141.266.76.3
There is no iptables enabled in our linux machine but i do not know how to check if port UDP:123 is open?



Any suggestion of help will be appreciated!
Thanks in advance!

 

[elgrandeperro]

What does "/usr/sbin/ntpq -p" say?

 

[herzel]

Thanks elgrandeperro for your reply
Please see the output of the ntpq -p:

remote refid st t when poll reach delay offset jitter
==============================================================================
tlrd-dc1.telrad .LOCL. 1 u 50 64 377 0.160 876773. 37.636

 

[elgrandeperro]

To see if ntp is open, from another host run:
ntpdate -q YOURNTPSERVER

This just queries the server for it's date.

 

[herzel]

Hi elgrandeperro
I actually now syncing our linux servers via ntpdate xxx.xxx.xxx.xxx (ntp server)of course after bringing down ntpd service .
so looks like ntp is open .

 

[elgrandeperro]

Do you have or remove these lines?

server 127.127.1.0 # local clock
fudge 127.127.1.0 stratum 10

 

[elgrandeperro]

I am a bit confused on why your time sources don't appear in ntpq -p. First, check to see if the default ntp.conf is actually /etc/ntp.conf. Check /etc/init.d/ntpd and /etc/sysconfig/ntpd

 

[herzel]

Thanks elgrandeperro
Actually the ntp server does list on the ntpq -p command, the printed name is the dns name of the ntp server.
for the first question there is no any localhost in the ntp.conf
and also i checked the other files nothing special.

 

[elgrandeperro]

The 127.127.1.0 is the local clock. This is only used if the other lower stratum clocks are unavailable (hence stratum 10).

Here is a very good writeup:

http://en.linuxreviews.org/NTP_-_Howto_make_the_clock_show_the_correct_time

 

[herzel]

Thanks elgrandeperro
I managed to put ntpdate at crontab every hour.
seems like for now its satisfy our needs .

Thanks very much for your replys and help.