atcomsystems.ca/forum Forums TDM Business Telephone Systems Avaya - Lucent Telephones & Systems ODD problem-Partner ACS line ports get screwy when exten ports connected to computer?

Greetings. I'm new to this forum, and I have tried searching, so apologies if this is a dumb question or I am posting it incorrectly.

I am also mostly a computer/voip guy, so if I get something wrong, let me know

Anyway I have a client who has a Partner system (installed by some other company that since went under). Processor is a Partner ACS R6, with two 206e (expansion) modules, R4.1. The system has 6 lines and about 20 or so extensions, all Partner 18D system phones. Voicemail is provided by a PartnerMail VS module, R5. This all sits in a 5 slot chassis, in the following order:
206e-206e-partnerACS-partnermailVS-nothing-nothing.
It is wired up through a power conditioner and a UPS backup. It all works great, and has been working for several years.

Current programming has lines 1,2,4,5,6 going into an Auto Attendant on the PartnerMail, with line 3 ringing a handful of phones for a while and then going to VMS.

The project is to replace PartnerMail VS with a computer-based telephony server (Asterisk on Linux). Partner communicates with its (or any other) voicemail system using a series of DTMF codes, which were mapped out by another guy doing the same thing. If interested, find that here . This will add a voicemail-to-email gateway, as well as future VoIP expansion.

The server has a Sangoma A200 series interface card, which has 4 FXO (line) ports.
I first backed up the PartnerACS using a Backup/Restore PC card. I then plugged the 4 line ports into 4 free extensions (26-29), and ensured that I could dial those ports and the system would pick up. I then assigned them to Hunt Group 7 and removed the PartnerMail ports (34-35). I then tried a few things, IE checking voicemail, etc. That worked fine, the Partner sent the codes and the server responded as I had programmed.

That's where the problem started

When a call comes in, if it is not picked up IMMEDIATELY (within 1-2 seconds), then ALL the external lines light up for a second (not just the one that a call came in on). They then all go out and the call is dropped. Dialing the server via the Intercom still works great, and it answers.

I switched Hunt 7 back to the partnermail. Still can dial server via intercom, although i now must punch in the code manually. Partnermail works.
But
Calls coming in on the AA lines get picked up quickly, and all the other line lights still come on for a second. Line 3 (the non AA line) gets answered and immediately dropped, with all the lights on.

After some troubleshooting, I found the following:
If I plug the server into the partner, on any extension port or combination of ports, on either of the 206e modules, the system enters this odd state where it is unable to correctly answer a call, and will not get out of it until I disconnect ALL the server lines and reset the partner system (power cycle or reset-save programming).

I have two possible theories-
1. however the Partner is sending data to its phones (data being display, buttons, lights, etc) is somehow getting an odd signal from the server,
2. There is some kind of electrical issue over the phone line. Both the Partner and the server are plugged in through the same outlet, same UPS, thus same ground so this should not matter...

So I am basically hoping that somebody can help me out. I am considering buying another brand of interface card (Digium TDM400) to see if that will help...

Thanks for any assistance!

It's actually a lot simpler than all that.

When you removed the VS mail's ports from Hunt Group 7, you left the ports with lines assigned and immediate ring. All 4 of them are trying to answer incoming calls.

You either need to physically remove the voice mail module from the carrier, or remove all the lines that are assigned to to voice mail's ports.

Let us know the progress of your project, I'm sure there are many that are interested

Thanks for the quick answer!

hmmm, that is an interesting idea. Not sure if its the right one though, as the VS is only 2 lines and I am pretty sure more than 2 lines light up. I know my server isn't answering because i watched the Asterisk CLI (server status display) and no calls came in, it didn't ring enough for the server to even detect it...

I have one or two more thoughts on this but nothing concrete at the moment, so I will try pulling the VS module when I go back there next week.

Also I will keep you guys up to date as this goes along. When it is done I will be writing a full report which will be posted on voip-info.org (for any interested, voip-info is a wiki site containing (duh) information about VoIP... very in depth. Will provide a link here when that's up. Voip-info has a high focus on Asterisk (free/open-source soft-pbx for Linux), not much proprietary stuff (avaya etc) other than how to integrate it with Asterisk....

First I would make sure that the second pair of the four extension/VM port connections are not connected- pins 1&4. Use two conductor line cords to be on the safe side. Pins 1&4 would normally be power and data used by the phones and it's not necessary for what you are doing, you only need T&R. Shorting pair 1&4 will remove power from the extension port until the system is reset, shorting several at once could bring the system down. Have no idea what those cards are doing with pins 1&4.

If you are unsure of the DTMF data being passed on a port I would suggest a Digit Grabber bridged across the T&R of the port. This will show you what's going on.

Oh and for sure pull the Mail VS out and remove those ports from 505,7.

-Hal

You should also set the ports for your new machine to "no ring"
When a call comes in, it rings all of the ports. They all answer. Only the first to answer gets the call. The others get dial tone from going off-hook. Call should only be directed to ring the vmail ports through #206,#506,#507

To clarify, I meant the extension ports connected to your asterisk. Use centralized programming to do this (all green lights should be "fluttering" not solid or blinking)

Ummm, you may want to read up on Partner programming. Extensions assigned to 505 group 7 are in a hunt group, call comes into it from lines assigned to 206, group 7 and rings the first available port. There is no need, and as a matter of fact nothing will happen if you set those extensions to no-ring. It is overriden by assigning them to 505,7.

506 and 507 will only affect those lines and extension assigned to group 7.

You could assign your lines and VM extensions to a hunt group other than 7. Matter of fact some after market VM systems work that way. Difference with group 7 is that it will pass and receive DTMF information from and to a Partner VM system.

Why would you NOT want your VM extension ports to ring? It will not pick up.

-Hal

It worked! Not sure if it was the removing the lines from ringing those ports or redoing the wires with only the tip/ring, but it works! Still needs some programming on the asterisk side before it's 100% ready but the hard part is over.

I will provide a link to my final report when it's done!

System is up and running and the customer loves it! Messages are delivered via email and voicemail, MWI works, life is grand

As promised, here is my report on it:
Click Me!

Thanks for the help guys, if you ever need Asterisk help zap me a PM (it emails, right?) or PM me on the Asterisk forums (forums.digium.com)...

Thanks again!

Nice complete article!

I assume that you have now removed the VS Mail from the system? Please go back, change the line cords from the Partner to the Asterisk to 4-conductor, just to confirm that the original problem was caused by the data pair of the Partner stations hitting the Samgoma cards.