cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
4408
Views
0
Helpful
9
Replies

Incorrect Time on E20/C-Series

Patrick Sparkman
VIP Alumni
VIP Alumni

We have some E20s and C-Series units where the time is displaying incorrect.  All the C-Series are static IPs, the E20s are a mix of static/dynamic.

They are all using their default NTP settings:

Mode:  Auto

Address:  either 0.ntp.tandberg.com or 1.ntp.tandberg.com

As I understand it, they get their NTP information from DHCP, and if none is provided they go by what is in the address field.  Does this apply to systems with static addresses as well?  Are there preferred settings for both DHCP and static assigned systems?  Any ideas and additional information would be greatly appreciated, as we're seeing variations in times from just a few minutes to over a half hour.

Thanks! Patrick

1 Accepted Solution

Accepted Solutions

Hi!

If NTP is set to auto it would use if DHCP is set the scope for NTP which is returned by the DHCP server.

If either no scope is returned or you do not use dhcp it will use the configured NTP server.

In theory NTP could also be looked up via a SRV-DNS record, but I do not think I had seen that by

Cisco TP endpoints for now.

The ping test shows that you can not resolve the name to an IP.

If you use DHCP DNS servers will automatically assigned.If you use a static IP config you would

have to to define the DNS servers yourself.

xConfiguration Network 1 DNS Server 1 Address: "8.8.8.8"

8.8.8.8 is a google DNS, best to be replaced with the proper DNS server for your network.

In addition NTP and DNS traffic should also be allowed in your firewall.

You can also try to set this which sets the NTP mode to manual and the NTP address to

199.4.29.166 is one address the tandberg ntp pool returend, you can change it to a internal or external ip

which is known to work.

xconfiguration NetworkServices NTP Mode: Manual

xConfiguration NetworkServices NTP Address: 199.4.29.166

Please reboot the endpoint after you did a change to be sure it worked.

Please remember to rate helpful responses and identify

View solution in original post

9 Replies 9

Hi Patrick

Do you have internet access for your video-endpoint ?

May be FW block access for non-trusted ntp?

Do you can ping 0.ntp.tandberg.com or 1.ntp.tandberg.com?

Check that you specified for the your DNS server?

No need to use NTP by default. You can use any other. For example, the ntp.org or time.windows.com.

Also you can use  internal ntp from local LAN.

Br.Oleksandr

hdewulf
Level 1
Level 1

Hello Patrick

Can you ping both NTP server at Tandberg from the C-series ? You need to have internet access.

$ ping 0.ntp.tandberg.com

0.ntp.tandberg.com is alive

$ ping 1.ntp.tandberg.com

1.ntp.tandberg.com is alive

$

If you can ping the NTP server, can you also run this command. Incase you haven't connection to it, you will see that you run on local clock.

$ ntpq

ntpq> associations

ind assid status  conf reach auth condition  last_event cnt

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

  1 39573  963a   yes   yes  none  sys.peer    sys_peer  3

ntpq> quit

$ ntpdc -p

     remote           local      st poll reach  delay   offset    disp

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

*LOCAL(0)        127.0.0.1       10   64  377 0.00000  0.000000 0.03044

$ hwclock --show ; date ; date -u

Tue Apr 10 07:57:04 2012  -0.095849 seconds

Tue Apr 10 07:53:33 CEST 2012

Tue Apr 10 05:53:33 UTC 2012

$

/Hans

Thanks for the replys and the info!

Br.Oleksandr -

We our VCS and bridge on the same network, using one of these two ntp addresses, and they seem to keep their correct time.

Hans -

This is what I got when I did a ping command from a C-Series for both ntp addresses.

$ ping 0.ntp.tandberg.com

ping: unknown host 0.ntp.tandberg.com

I did a test, and it seems I can reach the TANDBERG ntp server from our server, which are located in another building, but I can't from my office.  However I can browse to the address via a web browser.

Can you please explain the local clock function?

I take it the E20/C-Series doesn't go by DHCP ntp servers if using a static IP, from the replys I've gotten so far?  If these systems are set to use a static address, will they contact DHCP for a list ntp servers on our network, or will I need to insert our ntp servers into each of our systems.  The reason we simply left them with the default ntp server, is because we have almost 60 systems.

Patrick

Hello Patrick ,

-The unix command hwclock access the hardware clock on the board [local], this clock can loses time at a certain rate. That's the reason to use the NTP server to sync up the clock/time correctly from stratum server.

You can do some adjustment on time , if you didn't have NTP enabled via the comand :

--hctosys     set the system time from the hardware clock

--adjust      adjust the rtc to account for systematic drift since

                the clock was last set or adjusted

[tandberg:~] $ hwclock  --hctosys

[tandberg:~] $

But this isn't recommended .

- Regarding DHCP and static IP addr. You can use static addr, or DHCP server which assign the C-series an IP address. Once you have IP connection, and you can reach the NTP server  , the time will be sync up , if no big time diff. exist on both, the time will be sync and updated.

/Hans

So in my case, we have systems using static IPs, will they contact DHCP to obtain an ntp server address to contact, or will they go off of the manual address in the codec?  It's looking like I need to manually insert our local ntp address.

Or you use static IP addr,  or DHCP . You can't use both .

/Hans

Sorry, I guess I wasn't clear enough.  Because we're using static IPs, when in that configuration, they will use the ntp server that is listed in the address field?  It only contacts DHCP for an ntp server address ONLY if it's IP address is DHCP assigned?

Hi!

If NTP is set to auto it would use if DHCP is set the scope for NTP which is returned by the DHCP server.

If either no scope is returned or you do not use dhcp it will use the configured NTP server.

In theory NTP could also be looked up via a SRV-DNS record, but I do not think I had seen that by

Cisco TP endpoints for now.

The ping test shows that you can not resolve the name to an IP.

If you use DHCP DNS servers will automatically assigned.If you use a static IP config you would

have to to define the DNS servers yourself.

xConfiguration Network 1 DNS Server 1 Address: "8.8.8.8"

8.8.8.8 is a google DNS, best to be replaced with the proper DNS server for your network.

In addition NTP and DNS traffic should also be allowed in your firewall.

You can also try to set this which sets the NTP mode to manual and the NTP address to

199.4.29.166 is one address the tandberg ntp pool returend, you can change it to a internal or external ip

which is known to work.

xconfiguration NetworkServices NTP Mode: Manual

xConfiguration NetworkServices NTP Address: 199.4.29.166

Please reboot the endpoint after you did a change to be sure it worked.

Please remember to rate helpful responses and identify

I think I got it work now, I left everything as it is, and put in our DNS server.  I'm getting reports that the time is now correct on the system.  In short, we were missing the DNS entry for our static systems, which I guess made it impossible to resolve/contact the TANDBERG ntp servers.

Getting Started

Find answers to your questions by entering keywords or phrases in the Search bar above. New here? Use these resources to familiarize yourself with the community: