[c-nsp] 2821 VWIC2-2MFT-T1/E1 clocking issue

[Jay Hennigan]

Robert Blayzor wrote:

> Are both of these E1 links going to different telecom providers or just 
> to one telecom provider and one to a private link of sorts?  If they're 
> both going to telecom providers, shouldn't their clock source be 
> relatively close?  I'm thinking if their a carrier they probably should 
> be synced at stratum 1 BITS source if not at the very least 3e for all 
> their SDH/SONET equipment...  I believe a E1/T1 only requires stratum 4, 
> maybe even 5.  I'd find it a bit strange if two actual telecom carriers 
> had that much difference in clock source that you were seeing slips.  
> The other questions is, are you sure you're getting good clocking from 
> both providers, have you tried one and not the other, etc?

Both are going to different providers.  I agree that they all *should* 
tie back to stratum 1 clocks and have minimal slips.  However, it isn't 
uncommon to find that carriers just clock internally off of a local mux 
somewhere for data circuits.  Absent a highly accurate clock on our 
side, it would be impossible to tell which carrier was off-frequency. 
They could both be.

Even if we could prove that one carrier was inaccurate, the effort of 
escalating the issue to someone within the organization who understands 
the issue, cares enough to do something about it, and has the authority 
to fix it isn't likely to be productive.  "I've got this one E-1 
customer who is complaining about slips because he's also connected to a 
competitor.  Nobody else has a problem.  He could fix it with a separate 
WIC card.  We should spend five figures on a highly accurate clock for 
this POP and cable it to everything with a serial interface", isn't 
likely to play well.

Dealing with many carriers, I've found that getting them to grasp the 
concept of clocking issues at all is often difficult.

In the TDM voice world the carriers usually get it right.  I run 
multiple long-distance and local PRIs from different providers to the 
same switch all the time and never have a problem.  I suspect that 
plesiosynchronous data circuits just aren't given the same attention.

This issue is likely the reason that Cisco gave us the "clock source 
line independent" option, which works just fine as long as you don't 
have to configure it via the interface you're trying to manipulate!

--
Jay Hennigan - CCIE #7880 - Network Engineering - jay at impulse.net
Impulse Internet Service  -  http://www.impulse.net/
Your local telephone and internet company - 805 884-6323 - WB6RDV