NTP sync problem with Subscriber

gbressanin · ‎09-09-2008

I have a cluster 6.1(2) with a publisher and a subscriber.

On the publisher the NTP is correctly synchronised with an external Catalyst 6500. The publisher should update the subscriber with the correct time but it doesn't do that.

On the publisher:

admin:utils ntp status

ntpd (pid 5654) is running...

remote refid st t when poll reach delay offset jitter

==============================================================================

127.127.1.0 127.127.1.0 10 l 42 64 377 0.000 0.000 0.008

*10.16.63.10 127.127.7.1 10 u 714 1024 377 71.184 -37.782 9.862

193.35.65.120 0.0.0.0 16 u - 256 0 0.000 0.000 4000.00

synchronised to NTP server (10.16.63.10) at stratum 11

time correct to within 37 ms

polling server every 1024 s

Current time in UTC is : Tue Sep 9 11:48:27 UTC 2008

Current time in CET is : Tue Sep 9 13:48:27 CEST 2008

On the subscriber:

admin:utils ntp status

ntpd (pid 25396) is running...

remote refid st t when poll reach delay offset jitter

==============================================================================

*127.127.1.0 127.127.1.0 10 l 62 64 377 0.000 0.000 0.008

synchronised to local net at stratum 11

time correct to within 11 ms

polling server every 64 s

Current time in UTC is : Tue Sep 9 11:49:36 UTC 2008

Current time in CET is : Tue Sep 9 13:49:36 CEST 2008

I done some change on ntp master on the catalyst to change the stratum ( >9 ), and I restarted on both servers the NTP ( utils ntp restart). This is sufficient or do I have to restart the servers?

Thank in advance for any help.

Regards

Giovanni

Warren Swingle · ‎09-09-2008

I've has intermittent success using 65xx as a NTP reference. According to TAC, devices will not sync when a NTP is referenced from most catalyst switches.

If you have a Cisco router (NTP must be in sync), try pointing NTP on the CM's to the router and see if it helps.

HTH

- Warren

gbressanin · ‎09-09-2008

Hi Warren,

thank you for your response. I configure on CM a router as NTP server but I have the same issue. On the router the NTP is in sync as you can see below :

voicegwmi1#sh ntp statu

Clock is synchronized, stratum 8, reference is 127.127.7.1

nominal freq is 250.0000 Hz, actual freq is 249.9954 Hz, precision is 2**18

reference time is CC71F839.DD8CAD1D (09:39:37.865 CET Wed Sep 10 2008)

clock offset is 0.0000 msec, root delay is 0.00 msec

root dispersion is 236352.25 msec, peer dispersion is 236352.25 msec

Regards

Giovanni

Warren Swingle · ‎09-10-2008

Try using a external NTP server, like a trusted NTP reference on the Internet. Looks like the router is sync'd up internal, I've seen issues with that before.

gbressanin · ‎09-10-2008

But if I using an external NTP the stratum value is lower than 8, so I don't match the requirement of Cisco than wants stratum 9 in the CallManager.

Regards

Giovanni

Warren Swingle · ‎09-10-2008

You could try dropping the stratum level of the router to something below 9, that might help. However, below is a setup I've got working.

Below is NTP information from a Call Manager cluster that is NTP sync'd to a IOS router. The IOS router is sync'd to NTP servers on the Internet. Let me know if this helps.

Router#show ntp associations

address ref clock st when poll reach delay offset disp

+~66.250.45.2 209.51.161.238 2 998 1024 377 0.000 2.048 18.683

*~64.73.32.134 65.212.71.102 2 546 1024 377 0.000 -1.509 18.685

+~69.16.152.68 209.81.9.7 2 549 1024 377 0.000 13.418 14.831

* sys.peer, # selected, + candidate, - outlyer, x falseticker, ~ configured

Router#

Rotuer#show run | include ntp

ntp server 66.250.45.2

ntp server 64.73.32.134 prefer

ntp server 69.16.152.68

Router#

******** Call Manager Subscriber NTP Info

admin:utils ntp status

ntpd (pid 8452) is running...

remote refid st t when poll reach delay offset jitter

==============================================================================

127.127.1.0 127.127.1.0 10 l 3 64 377 0.000 0.000 0.004

*10.10.128.20 10.10.128.10 4 u 82 128 377 0.172 -0.060 0.014

synchronised to NTP server (10.10.128.20) at stratum 5

time correct to within 102 ms

polling server every 64 s

Current time in UTC is : Wed Sep 10 16:08:57 UTC 2008

Current time in America/Chicago is : Wed Sep 10 11:08:57 CDT 2008

admin:

******** Call Manager Publisher NTP Info

admin:utils ntp status

ntpd (pid 8708) is running...

remote refid st t when poll reach delay offset jitter

==============================================================================

127.127.1.0 127.127.1.0 10 l 33 64 377 0.000 0.000 0.008

*10.10.128.10 64.73.32.134 3 u 602 1024 377 1.153 -0.118 0.035

synchronised to NTP server (10.10.128.10) at stratum 4

time correct to within 95 ms

polling server every 1024 s

Current time in UTC is : Wed Sep 10 16:12:05 UTC 2008

Current time in America/Chicago is : Wed Sep 10 11:12:05 CDT 2008

admin:

Warren Swingle · ‎09-10-2008

Forgot to mention IP's, from the post above...

Router: 10.10.128.10

Publisher: 10.10.128.20

Subscriber: 10.10.128.21

gbressanin · ‎09-11-2008

Hi,

I linked the NTP to an internet server but no changes, the issue is present again.

The router's config about NTP is like your config.

I think that I should take a ethernet trace , but this is a remote site and will do that in the next mouths.

Thank you

Regards

Giovanni

gbressanin · ‎09-11-2008

Hi,

I'm looking with the browser in the "Cisco Unified Reporting" page and under "Unified CM Data Summary" the value on the row "NTP server" is 0.

Can you check if also on your server is the same??

Thank you

Giovanni