NTP - Timestamp of Cisco 3560 Switch not synchronized with the configired NTP server

[Jayson Cruz]

Hello everyone,

Greetings! I would like to make an inquiry regarding ntp concepts.

I have a Cisco 3560 Switch that has an ntp configuration that makes the connected router its ntp server. I noticed that the timestamp on the cisco 3560x is different from time stamp on its server.

I am just wondering whether my configuration is correct or there are some concepts thats I am not aware yet with regards to ntp.

Below are the ntp configurations from the switch and the router which acts as a ntp server.

Cisco 3560

ntp clock-period 36027449
ntp source Vlan100
ntp server 10.42.92.254
end

C3560#show ntp associations

      address         ref clock     st  when  poll reach  delay  offset    disp
*~10.42.92.254     10.56.0.206       4    59    64  377     2.7   -0.62     0.2
 * master (synced), # master (unsynced), + selected, - candidate, ~ configured

C3560#show ntp status
Clock is synchronized, stratum 5, reference is 10.42.92.254
nominal freq is 119.2092 Hz, actual freq is 119.2137 Hz, precision is 2**18
reference time is DC969FC9.954CF283 (00:20:25.583 UTC Tue Apr 11 2017)
clock offset is 0.0988 msec, root delay is 28.46 msec
root dispersion is 81.60 msec, peer dispersion is 0.64 msec

Cisco 2900

ntp source Loopback0
ntp server 10.56.0.206 prefer
ntp server 10.63.255.226
end

C2900#show ntp associations
  address         ref clock       st   when   poll reach  delay  offset   disp
*~10.56.0.206     202.60.94.11     3    546   1024   377  5.941  -1.835 14.843
 ~10.63.255.226   .STEP.          16      -   1024     0  0.000   0.000 15937.
 * sys.peer, # selected, + candidate, - outlyer, x falseticker, ~ configured
  
C2900#show ntp status
Clock is synchronized, stratum 4, reference is 10.56.0.206   
nominal freq is 250.0014 Hz, actual freq is 249.9749 Hz, precision is 2**16
reference time is DC969D6C.AFF6D343 (10:10:20.687 AEST Tue Apr 11 2017)
clock offset is -1.8359 msec, root delay is 27.10 msec
root dispersion is 80.09 msec, peer dispersion is 14.84 msec
loopfilter state is 'CTRL' (Normal Controlled Loop), drift is 0.000106050 s/s
system poll interval is 1024, last update was 553 sec ago.

[lpassmore]

Hi Jayson

The config looks just fine and it appears to be synchronized correctly.

How different are the times between the devices?  

I would also suggest the timezone is not set on your switch.  Does that account for the difference?

If the times were different by a large amount (not related to timezone differences), NTP can take a while to adjust.  Maybe wait a day or so and see if fixes itself

[Jayson Cruz]

Hi Ipassmore,

Thank you very much for taking time to read and comment on my inquiry.

Regarding you questions, time difference is around 9 hrs. As per the timezone both devices are on the same location.

For the configuration, the ntp config was in place more than a year already as this is an old infrastructure.

Thank you very much!

[lpassmore]

If the time difference is around 9 hours then timezone configuration is most likely the issue.  Check that you have the appropriate 'clock timezone' command in the config

Also, you have not really said which timestamp is different.  Look at differences between the devices in the 'service timestamp' commands

If it is the actual clock that is out and it is out by more than 5 minutes then delete the ntp server commands, turn on debug ntp all and then redo the ntp server 10.42.92.254 command and post the results from the debug and the show ntp status and show ntp association commands

Cheers

LP