corruption?

[fishobx]

First of all, this is my first post and I am so thrilled to find this forum!!!! I've worked on Avaya products for about 6 years now and hope I can at some point share a fraction of the knowledge that I have absorbed in the last week or so of reading/lurking.

I have a plastic R3 magix (Legend upgrade, replaced the processor and an 008 ATL with an 008 MLX) with wierd problems. First question, should I have re-programmed from scratch instead of just doing a conversion followed by a board renumber, renumbering, re-programming buttons on stations affected by replacing the 008? that card was in about slot 7 or so and followed by an 012 and an 008 MLX, so some stations were affected by the new adjunct extensions. I've received mixed information, even from Avaya directly, about whether or not to manually reprogram in similar situations, and I did have a mess with corrupted software after replacing a BRI card with PRI in slot 1 once upon a time....ended up re-programming a full 3 cabinet system manually.

Anyway, the problems we are having are *not* exclusive to the stations affected by replacing that card. In fact, they seem to be worse for stations on lower logical ids. Basically, a user on an outside call will get interrupted by a new incoming call. The original caller and the outside caller can usually hear the conversation between the operator and the new outside caller, but the new caller and the operator cannot hear the original participants, and the original participants can no longer hear each other. In at least one instance, the original callers held on until the new conversation was over, and they were then "reconnected" and compared notes on what happened. Usually the original outside caller hangs up, and it is assumed they heard nothing, as if the call was dropped. At least one caller called back and told the original user that this was the case, but I assume that sometimes the caller does hear the new call but hangs up when it is obvious that no one can hear them. This has happened sometimes 2 or 3 times a day, then not again for a week or so, intermittent since the upgrade 2 or 3 months ago.

The customer did say they had lots of "wierd" problems prior to the upgrade. The system administrator had not tracked problem details, just told users it will 'get better' after the upgrade. (We also replaced their old external voice mail w/ the new Messaging, and apparently most of the problems were voice mail related). Customer cannot confirm whether this *exact* problem occured prior to the upgrade.

Techs have been onsite and inspected wiring and affected stations and found nothing, and haven't been able to recreate the problem, even after multiple calls to every line in switch. I've looked at the programming and there are no shared system access buttons or anything else I can see that might be contributing. We called Avaya this week and they said it was either a bad processor or corrupted software. I tend to believe the problems were occuring with the previous processor, but I guess we will try another processor just to be sure before re-programming.

A few more details, we have not been able to isolate this to a particular line or lines. It usually seems to involve the operators (4 DLCs that answer about 5 incoming LS lines) though occasionally it is with another extension that has a personal LS line. Also, one report of an individual's voice mail actually recording a conversation between an operator and an outside caller--caller was not calling the extension that had the wierd vmail message, and apparently is was a pretty long recording.

Wanted to see if anyone had any ideas/suggestions before I start completely reprogramming this switch....

thanks!

 

[merlinman]

I have a few ideas.

But until you get to the bottom of this, don't talk outloud about these. (You will understand more as I go on.)

First, mixing up modules, as you did when you converted this system will not cause cross talk.

Mixing up modules will cause features for extensions to get scrambled, and other things. That is, say you have logical id 45, and it's originally an ATL set. (The old Membrane type)

You then change it to be an MLX - well, you just added logical ID 45 and 46 (in simplistic terms) That is, where the ATL set had only 1 extension associated with it, the MLX now has an extension and an Adjunct. That additional ID can cause extensions BEYOND that address to take on the programming of what was (or was not) on one of the ATL sets. Again, this is an oversimplification, but 1 affect of mixing modules is assigning extension programming to places you did not expect it to be.

Another affect is how ASSOCIATIONS of the LOWER EXTENSIONS are tied into the HIGHER EXTENSIONS. Again, you now have more ID's in the system, so if ID 46 was in a calling group, and it was an ATL set, then ID 46, which now may be an adjunct, will be in that calling group and the OLD ASSOCIATION has changed. Again, this is an oversimplification.

NOW TO YOUR PROBLEM:

I doubt that CORRUPTED SOFTWARE will cause cross talk.

Did you get that theory from a TIER 3 Engineer, or an "UP FRONT" person. (If you have access to AVAYA, perhaps you would want to ask for a FIELD ENGINEER...)

But, moving on....

If I were to try to isolate the problem to a trunk I would do 2 things:

1) Setup a PRINTER (LAPTOP) to monitor SMDR.

2) Instruct the USERS to LOG THE TIME OF EACH FAILURE (and of course, their extension).

But, I don't think you have a problem with the lines.

I think this problem cause would be EASILY FOUND if you could get a FIELD ENGINEER on site to MONITOR all Extensions that Answer and distribute calls.

That is, are all calls to this system aswered by an ATTENDANT? (You said MOSTLY)

How many ATTENDANTS do you have? (You said 4)

Remember, ALL ATTENDANTS CAN CAMP ON CALLS.

Read what you wrote, and look at it in a different light, based on the facts I just stated.

If you read between the lines here, you will understand why I suggest you pursue this in a rather quiet manner.

OK, so I've said a bunch, and only a little may be on target, but I have given you a few tools to start with, if nothing else.

 

[fishobx]

Thanks, merlinman. You did give me some things to think about, but I have to admit I'm still scratching my head, trying to read between the lines. I understand exactly what you mean about replacing the card, and did indeed have to correct programming on extensions as far as button programming, extensions in cover group 30, etc. I tend to think this was an existing problem and not caused by our upgrade.

I didn't talk to Avaya directly--one of our technicians called, and I don't know who he talked to. Probably an 'up front' person. I'll give them another call and ask them to monitor...

I re-read the feature description for camp on, which I seldom run into since most systems have voice mail, but it's not clicking for me. Any other "hints" would be greatly appreciated!

thanks again for your quick reply!

 

[exsmogger]

Is your processor a CKE4? It should say on the label. If you have a CKE3 I think that's your problem. Also, are your LS lines on an old 800 instead of a 800 GS/LS? I've had problems using old 800 mods on Legend/Magix.

Brian Cox

http://www.jandjcommunications.com

 

[fishobx]

It's a new processor. I'll double check the line cards though.

thanks!

 

[merlinman]

I would have to ask again, do you have access to Avaya Tech Support, as in being an Authorized Dealer?

If so, I again would advise getting their TIER 3 Engineers involved, and perhaps even a Field Engineer. (If they still have those)

They can do all the tricks I can and, if you are authorized, it (thier support) comes with the territory.

merlinman@comcast.net

 

[fishobx]

Yes, I have access, and I am going to try them again, but as I said, they told my technician to try replacing the processor and/or reprogram manually, suspecting software corruption. I'm not going to give up yet. I'll request a tier 3 engineer.

thanks again!