[comp.dcom.telecom] Help with Rotored Lines/ Rack Mounted Modems

Pushpendra Mohta <pushp@cerf.net> (07/14/90)

I recently acquired a set of Telebit2500 Rack Mount Modems and found
what I consider a major irritant.

Unlike the stand-alones there is no switch on the control panel of the
rackmounts to make a particukar modem go busy or off hook.  (You can
connect to the modem in AT mode and then make it go offhook but my
modems will be all over California and technical help may not be
available at all sites )

Ordinarily this would not be a problem, but the application I have is
a dial up terminal server with the phone lines on a rotor.  If the
first modem is busy , the call forwards to the next one and so on.
Should the modem or the terminal server port go bad and the modem does
not go off-hook, all ports beyond that one will not be utilized.

I wonder if there are feautures available in Pac-Bell territory on the
ROTORS which in effect will allow forward on busy AND forward if a
particular line does not answer in say four rings simultaneously. I
tried talking to the rep here without a clear answer.

Are there any other suggestions ? 

Thanks,

pushpendra
CERFNet 

faunt@cisco.com (Doug Faunt N6TQS 415-688-8269) (07/15/90)

It turns out, that the way to busy-out a bad modem in a T2500
rack-mount system is to pull the modem out of the rack, and send it in
to get it fixed, presumably.  This can be done on a running card cage
with no problems.  If you could get someone to flip a switch, you
could get them to pull a card.

Another solution is to check out various styles of "call
distribution".  We have a ATT Systme 75 here at cisco, and one of the
styles of call distribution is called "uniform call distribution".  It
tries to route the next call to a group to the least-used line
available, but never the same line as the last call, so if there's a
bad modem, the user just hangs up and redials, and gets a different
line.  Bad modems are indicated pretty reliably by various statistics
that our, cisco, terminal servers keep.  I don't know if "UCD" is
available from central offices, however.  good luck, faunt@cisco.com

julian@bongo.uucp (Julian Macassey) (07/19/90)

In article <9796@accuvax.nwu.edu>, faunt@cisco.com (Doug Faunt N6TQS
415-688-8269) writes:
 
> It turns out, that the way to busy-out a bad modem in a T2500

The simple way is short the damn phone line. This can be done with
high tech paper clips or fancy plugs across the 66 block. This is a
no-fail guaranteed way to do it. Funny I had to explain this to
"Sprint PC-Pursuit" last week. But they are a telephone company, so
how are they to know how to busy out a modem (-: I am always amazed
how ignorant and incompetant some employees are.

Anyhow a good administarator of telco gear should check all trunks and
associated equipment on a regular basis. This is a task an operator
can do first thing in the morning. It is also handy to check 800
numbers on thge local loops as well, they do die sometimes.

> Another solution is to check out various styles of "call
> distribution".  We have a ATT Systme 75 here at cisco, and one of the
> styles of call distribution is called "uniform call distribution".  It
> tries to route the next call to a group to the least-used line
> available, but never the same line as the last call, so if there's a
> bad modem, the user just hangs up and redials, and gets a different
> line.  Bad modems are indicated pretty reliably by various statistics
> that our, cisco, terminal servers keep.  I don't know if "UCD" is
> available from central offices, however.  good luck, faunt@cisco.com

Yes, UCD is certainly available on the 5ESS switch. See Part 2 of "5
ESS Switch The Premier Solution - Feature Handbook" Number 235-390-500. 
But you may not want to pay for it. It should be cheaper to check the
trunks and busy out the bad ones until the modem or loop is fixed.

When in doubt - check it out.


Julian Macassey, n6are  julian@bongo.info.com  ucla-an!denwa!bongo!julian
N6ARE@K6IYK (Packet Radio) n6are.ampr.org [44.16.0.81] voice (213) 653-4495

john@bovine.ati.com (John Higdon) (07/19/90)

Julian Macassey <julian@bongo.uucp> writes:

> Anyhow a good administarator of telco gear should check all trunks and
> associated equipment on a regular basis. This is a task an operator
> can do first thing in the morning. It is also handy to check 800
> numbers on thge local loops as well, they do die sometimes.

I'll second that. Even here at home, where I have five modems working
various trunks, in and out through the PBX, a weekly check is
performed. From a phone, I dial the extension number of each modem and
check for answer, and then using one of the electronic phones, punch
up each CO line that is involved with modem use and check for dial
tone.

This may seem silly, but this routine check solved a problem that had
gone unnoticed. One of my uucp neighbors calls on my 1200 bps modem at
night because they run out of Telebits due to heavy traffic. As a
courtesy, they don't tie up Telebits with a "slow" call. Somehow, I
had kicked the modular cord out of the modem, and it went out of
service.  All week, I noticed that my mailing list subscriptions were
coming in late but passed it off as just "net aberations". When I did
my weekly check, the 1200 bps modem failed to respond and the problem
was readily corrected. Suddenly, my Telecom Digests started arriving
on time once again.


        John Higdon         |   P. O. Box 7648   |   +1 408 723 1395
    john@bovine.ati.com     | San Jose, CA 95150 |       M o o !