Subject: Re: Clock synchronization with ISDN
To: None <tech-kern@NetBSD.ORG>
From: Jan-Hinrich Fessel <oskar@unna.ping.de>
List: tech-kern
Date: 02/18/1997 08:27:45
--==_Exmh_-1660047120P
Content-Type: text/plain; charset=us-ascii

In message <199702180234.TAA19655@continuum.epoch.org>you write:
> On Mon, 17 Feb 1997 11:53:59 -0800, Jonathan Stone  
> Most of the timestamps present in ISDN are generated by the last switch  
> before your end of the BRI or PRI, therefore latency will be reasonably low  
> (not more than your ISDN link itself (round-trip times of no more than 2-3ms  
> worst case).

But the fact is, the last switch doesn't alway generate acurate stamps.
Mine (a SIEMENS built AFAIK) is constantly behind DCF-77 syncronized time.  
And the time delay varies from 1 to 2 seconds.
I do not find this to be accurate enough, but it's still better than my local 
timeclock on one of my x86 boxes.
 
Even my PBX on the other channels does better with a Quartz-Oscillator...

I have heard of switches that are DCF-syncronized, but you can't count on this.

I have noticed the timestamps are more behind real time when they come near to 
"no B-Channels availible" messages, It looks like the switch in question 
looses clock ticks when under heavy load.

Gruesse
       Oskar

 ==============================================================================
 					Tragbar ist, was nicht herunterfaellt.
						    oskar@zappa.unna.ping.de
 ==============================================================================



--==_Exmh_-1660047120P
Content-Type: application/pgp-signature

-----BEGIN PGP MESSAGE-----
Version: 2.6.3i

iQCVAwUBMwlZ8LA63tqjeAhZAQHvowQAmSjFadl34Mus+VaWFXqphZdlG4uVKsIw
jHRyRrtoduTpMz1giORTtAQK1rRj1w3baWwEJLBFOmJj/6hYwLRKX1XNZ3a9FaG7
f/a+vhXSjx5fL3+Z+2dbUZ41DKVMklsQ2/9VIKQj3kq/ssyWIN2/ya9tqwLGraGW
8I0IchD/qXg=
=r3CL
-----END PGP MESSAGE-----

--==_Exmh_-1660047120P--