NTP Issues

zaeem · 05-13-2011, 07:34 AM

Dear Guys,

I have been working on NTP to find out resolution of my issues unable to find. let me briefly explain here. I have three servers and no server is fully synced with remote NTP server.

Server 1:
/etc/ntp.conf is as follows;

Quote:

restrict default kod nomodify notrap nopeer noquery
restrict -6 default kod nomodify notrap nopeer noquery
restrict 127.0.0.1
restrict -6 ::1
server 192.168.0.12
server 192.168.0.13
driftfile /var/lib/ntp/drift
logfile /var/log/ntp.log
keys /etc/ntp/keys

NTP log shows following

Quote:

13 May 15:46:43 ntpd[3417]: synchronized to 192.168.0.12, stratum 1
13 May 15:46:48 ntpd[3417]: time reset +5.334444 s
13 May 15:46:48 ntpd[3417]: kernel time sync disabled 0001
13 May 15:50:43 ntpd[3417]: synchronized to 192.168.0.12, stratum 1
13 May 15:51:49 ntpd[3417]: no servers reachable

Server 2
/etc/ntp.conf has following entries

Quote:

restrict default kod nomodify notrap nopeer noquery
restrict -6 default kod nomodify notrap nopeer noquery
restrict 127.0.0.1
restrict -6 ::1
server 192.168.0.12
server 192.168.0.13
server 127.127.1.0 # local clock
fudge 127.127.1.0 stratum 10
driftfile /var/lib/ntp/drift
logfile /var/log/ntp.log
keys /etc/ntp/keys

log file shows

Quote:

13 May 16:57:29 ntpd[10020]: synchronized to 192.168.0.12, stratum 1
13 May 16:58:35 ntpd[10020]: synchronized to LOCAL(0), stratum 10
13 May 16:59:40 ntpd[10020]: synchronized to 192.168.0.12, stratum 1
13 May 17:00:43 ntpd[10020]: synchronized to LOCAL(0), stratum 10
13 May 17:00:46 ntpd[10020]: synchronized to 192.168.0.12, stratum 1
13 May 17:01:49 ntpd[10020]: synchronized to LOCAL(0), stratum 10
13 May 17:02:53 ntpd[10020]: synchronized to 192.168.0.12, stratum 1
13 May 17:03:58 ntpd[10020]: synchronized to LOCAL(0), stratum 10

Server 3
/etc/ntp.conf has following entries

Quote:

restrict default kod nomodify notrap nopeer noquery
restrict -6 default kod nomodify notrap nopeer noquery
restrict 127.0.0.1
restrict -6 ::1
server 192.168.0.12
server 192.168.0.13
server 127.127.1.0 # local clock
fudge 127.127.1.0 stratum 10
driftfile /var/lib/ntp/drift
logfile /var/log/ntp.log
keys /etc/ntp/keys

log file shows

Quote:

13 May 16:53:26 ntpd[4412]: synchronized to 192.168.0.12, stratum 1
13 May 16:54:31 ntpd[4412]: synchronized to LOCAL(0), stratum 10
13 May 16:55:36 ntpd[4412]: synchronized to 192.168.0.12, stratum 1
13 May 16:56:40 ntpd[4412]: synchronized to LOCAL(0), stratum 10
13 May 16:57:46 ntpd[4412]: synchronized to 192.168.0.12, stratum 1
13 May 16:58:52 ntpd[4412]: synchronized to LOCAL(0), stratum 10
13 May 16:59:56 ntpd[4412]: synchronized to 192.168.0.12, stratum 1
13 May 17:01:02 ntpd[4412]: synchronized to LOCAL(0), stratum 10
13 May 17:02:08 ntpd[4412]: synchronized to 192.168.0.12, stratum 1

I don't know why it sync time alternatively with remote NTP server and LOCAL whereas there is not issue in connectivity/reachability of NTP server and NTP clients. Also server 1 reporting kernel time sync disabled 0001. Please help me to resolve these issues.

Note: All servers has CentOS 5.5 installed.

ncsuapex · 05-13-2011, 07:57 AM

With ntpd running what does this command show?

Quote:

ntpq -p

zaeem · 05-13-2011, 09:02 AM

Dear ncsuapex,

Thanks for reply. Following is the information you asked.

Server 1

Quote:

# ntpq -p
remote refid st t when poll reach delay offset jitter
==============================================================================
192.168.0.12 .GPS. 1 u 49 64 377 1.332 457.550 20.541
192.168.0.13 LOCAL(0) 11 u 61 64 377 16.983 -944.08 20.403

Server 2

Quote:

# ntpq -p
remote refid st t when poll reach delay offset jitter
==============================================================================
192.168.0.12 .GPS. 1 u 119 1024 377 1.711 82.611 19.481
192.168.0.13 LOCAL(0) 11 u 534 1024 377 17.525 -1306.5 19.427
*LOCAL(0) .LOCL. 10 l 12 64 377 0.000 0.000 0.001

Server 3

Quote:

# ntpq -p
remote refid st t when poll reach delay offset jitter
==============================================================================
192.168.0.12 .GPS. 1 u 337 1024 377 1.555 137883. 162.458
192.168.0.13 LOCAL(0) 11 u 691 1024 377 17.235 136426. 162.288
*LOCAL(0) .LOCL. 10 l 64 64 377 0.000 0.000 0.001

michaelk · 05-13-2011, 10:54 AM

Post the ntp.conf files for your time servers. 192.168.0.12 and 192.168.0.13

zaeem · 05-14-2011, 02:57 AM

Dear michaelk,

I don't have access to .12 and .13 time servers. However I will try to get conf files.

michaelk · 05-15-2011, 08:32 PM

Are these real or virtual machines?
For some reason the ntp servers have a offset and jitter are high and so ntp uses the local clock as the best source. Server 1 is not configured to use the local clock.

You could try

Code:

server 192.168.0.12 burst iburst
server 192.168.0.13 burst iburst

ncsuapex · 05-17-2011, 08:51 AM

If the offset is that high NTP has problems syncing. Try stopping ntp then running:

Quote:

ntpdate 192.168.0.12

Then restarting ntp and running the ntp -q command again. Also check /var/log/messages and /var/log/ntp.log