Solved: network-clock-participate

liettig · ‎08-17-2009

Hi,

is there anyone who can kindly show me when to use the network-clock-participate command?

Is it needed in a router with 2 VWIC BRI (2 ports) connected to the same provider?

Regards

paolo bevilacqua · ‎08-17-2009

As well network-clock-select, it's needed if you do fax or modem. Otherwise, it's recommended to have but no problems should occur if you don't.

View solution in original post

Nicholas Matthews · ‎08-18-2009

If there are slips the best idea is to start doing loopback tests on the links. Make a T1 loopback cable (1-4, 2-5, hairpinned) and see if any of the sides are getting slips.

In your 4 examples, if all 4 BRI ports go to the same clocking domain, they will be equivalent. Many people like to use a preference of 2 in case one goes down as well.

-nick

View solution in original post

paolo bevilacqua · ‎08-17-2009

As well network-clock-select, it's needed if you do fax or modem. Otherwise, it's recommended to have but no problems should occur if you don't.

Nicholas Matthews · ‎08-17-2009

Each set of DSPs have one clock. By default, the DSPs will be sync'd to the internal self-created clock.

Even when you configured 'clock source line', the DSPs choose the internal clock first.

If you have 2 PRIs for example, one to two different providers, you will be connected to two different 'clocking domains'. This command allows you to choose which timing you would like to sync to.

As long as both of your BRIs are on the same clocking domain, you should be fine. Use the network-clock-select command to choose your provider's clock for the DSPs to sync to.

If you do have slips, you can have slight clicks and pops on the line, as well as fax failures.

hth,

nick

liettig · ‎08-18-2009

Hi,

thanks for the reply.

So, if I have 2 VIC2-2BRI-NT/TE modules in the same router (interface BRI0/0/0

interface BRI0/0/1 interface BRI0/1/0 and

interface BRI0/1/1) all connected to the same provider, the following configurations are correct:

network-clock-participate wic 0

network-clock-participate wic 1

network-clock-select 1 BRI0/0/0

that is the same of

network-clock-participate wic 0

network-clock-participate wic 1

network-clock-select 1 BRI0/0/1

that is the same of

network-clock-participate wic 0

network-clock-participate wic 1

network-clock-select 1 BRI0/1/0

that is the same of

network-clock-participate wic 0

network-clock-participate wic 1

network-clock-select 1 BRI0/1/1

is it correct?

nick, is there a way to debug if there are slips?

Nicholas Matthews · ‎08-18-2009

If there are slips the best idea is to start doing loopback tests on the links. Make a T1 loopback cable (1-4, 2-5, hairpinned) and see if any of the sides are getting slips.

In your 4 examples, if all 4 BRI ports go to the same clocking domain, they will be equivalent. Many people like to use a preference of 2 in case one goes down as well.

-nick

paolo bevilacqua · ‎08-18-2009

You cannot do loopback tests on a BRI interface, neither in software nor using a crossed cable.

However, on BRI slips will show as CRC and other errors on the D-channel /0/x/y:0 consequently it's always recommended to use network-clock-select. The drawback is that will clutter your log a little.

Again I'd like to remind the OP to please rate useful posts with the scrollbox below.

liettig · ‎08-18-2009

thank you for your answers!

paolo bevilacqua · ‎08-18-2009

You're welcome, thank you for the nice rating and good luck.