NTP PROBLEM

[mahalingam1]

Hi all!

I have the following problem whith NTP in a sunfire 15K under solaris9. cheking the /var/adm/messages file i can see the following messages, these messages appears a lot of times:

Oct 2 10:59:05 machine xntpd[720]: [ID 774427 daemon.notice] time reset (step) 102.837587 s
Oct 2 11:02:30 machine xntpd[720]: [ID 774427 daemon.notice] time reset (step) -102.952417 s
Oct 2 11:10:25 machine xntpd[720]: [ID 774427 daemon.notice] time reset (step) 104.858177 s
Oct 2 11:13:50 machine xntpd[720]: [ID 774427 daemon.notice] time reset (step) -102.994298 s


I was cheking the ntp.conf and this looks fine

more /var/ntp/ntp.drift
96.503

It's called me the attention the correction time is the same but with opposite sign

Please anyone tell me what's happen?

Thanks in advance and regards

 

[elgrandeperro]

ntpq -p ? what is the jitter?

 

[Annihilannic]

I have seen something similar in the past when I was pointing at two NTP servers whose times were unfortunately quite different...

Annihilannic.

 

[elgrandeperro]

Exactly. You probably need to disable the one or fix the one with higher jitter.

 

[mahalingam1]

hi elgrandeperro and Annihilannic

I saw yours answers today. I don't recive the email notification about it

When i restart the ntp client the stratum of the machine is 2. Acording with our ntp topology, but when the xntpd is up for a few time the ntptrace show stratum 16 and insane condition:

ntptrace
localhost: stratum 16, offset 0.000040, synch distance 106.24739
dcimser01.dimserso.int: stratum 2, offset -0.007769, synch distance 0.10275


ntpq -p
remote refid st t when poll reach delay offset disp
==============================================================================
prefer_server. .LOCL. 1 - 8h 64 0 0.61 2361.61 16000.0
secundary server tock.usno.navy. 2 - 273 64 0 1.25 -101820 16000.0



ntpq> as
ind assID status conf reach auth condition last_event cnt
===========================================================
1 60876 90f4 yes yes none insane reachable 15
2 60877 90f4 yes yes none insane reachable 15

Any ideas why the xntp becomes a stratum 16 from stratum2?


Thanks in advance and resgards

 

[mahalingam1]

I was checking another ntp clients that work fine and the output of the ntpq -p is:

ntpq -p
remote refid st t when poll reach delay offset disp
==============================================================================
#10.17.1.50 .LOCL. 1 u 493 1024 377 0.60 -46.387 8.26
+10.17.1.120 192.5.41.41 2 u 649 1024 377 0.50 -104256 11.93


as you can see in the first column appear the "#" sing: Selected for synchronization; but distance exceeds maximum. In the machine that not work fine this sing doesn't appear.

Any ideas???

Thanks

 

[mahalingam1]

Hi all i resolved the problem deleting the server line in the /etc/inet/ntp.conf.

Regards

 

[Annihilannic]

Doesn't that mean that NTP is no longer synchronising with anyone? In that case you may as well just stop the service...?

Annihilannic.

 

[mahalingam1]

Hi Annihilannic.In the ntp.conf file only appear the line about the server prefer.

Now the stratum is 2,

ntptrace
localhost: stratum 2, offset 0.000072, synch distance 10.40128
acebo.dimserso.int: stratum 1, offset -0.001657, synch distance 10.39061, refid 'LOCL'

Mahalingam