Cell Relay Archive

[mrand@my-deja.com]

In article <y5Yd5.2869$J7.30756@grover.nit.gwu.edu>,
  <agallo@foghorn.nit.gwu.edu> wrote:
> Thanks for all the replies- I've learned a great deal about SRTS
lately!
> First, I do mean ATM over DS-3.  It is one of multiple PNNI links on
our network.
> We run both AAL5 (for LANE) and AAL1 (for PBX traffic- we've got a
Lucent G3 running
> in PNC or native ATM mode- no circuit emulation). Also, I did turn on
DS-3 scrambling
> with no improvement.  I've tried every combination of mode (hcs vs.
plcp) and framing
> (cchannel vs. cbit) on the ASX-200BX- to improvement.
>
> The radio link by itself tests great with a TBERD (better than 1x10-9
BER) and
> the ATM switches test fine in loopback.  The two systems together,
however, don't
> work well.

The above sounds like a terminal loopback (which better work!),
and that isn't what you really need.

In attempt to be complete, I may state some obvious debugging
ideas... please excuse those ideas if you've already tried them.

Have you tried a facility loopback (sometimes called link or line
loopback) on each of the switches (one at a time)?  IE,
loopback one ATM switch TOWARDS the radio link.
This will help verify that the DS-3 layer is really working
properly.

The possiblities as I see them are:
1. The DS-3 interface on the radio link doesn't work properly.
2. The DS-3 interface on the switch doesn't work properly.
3. The radio and/or switch aren't configured correctly(identically?)

You've basicly knocked item #1 out with the TBERD (you can
note all the config settings and possibly help knock out item #3).

You can check #2 a few different ways.  The first is as I
described above with the facility loopback.  If that does NOT
work, then take the radio link out of the picture: use a cable
to loop the switch back on itself.  If that doesn't work,
connect a DS-3 ATM test set to the switch and figure out which
direction it is that isn't working.

Part of item #3, as others have mentioned, might be a timing issue.
Have you inspected the timing configuration for the DS-3
closely, and/or tried changing any of the timing settings?

The thing I'd probably try first is to connect the TBERD as you
did in #1.  When you see the link working, connect the TBERD to
the switch withing changing anything on the TBERD.  The DS-3 layer
should be fine - you may not be able to pass data, but that's not
where your real problem is.

>  I'm getting a high number of DS-3 framing errors, P-bit errors, out
of
> frame, all eventually leading to ATM header checksum erros; from time
to time
> this is enough to bring down ATM signalling between the switches
cuasing the
> SVCs to reroute to a backup T1.
>
> At this point, the radio vendor is saying that SRTS would be needed.
As I'm sure
> some of you could tell by my question- I didn't know to much about
it!
> From the responses, this new "requirement" for SRTS does not sound
appropriate.

SRTS almost for sure is not the answer to your problem (not to mention
it is probably REALLY REALLY expensive).

   Marc


Sent via Deja.com http://www.deja.com/
Before you buy.