Magix keeps crashing after adding 100D

[mikedphones]

We have a magix/legend in the field carded with
0=CKE5 R4
1=408 MLX
2=800 GS LS
3=008 MLX
4=008 MLX
5=016 TTR

6=016 TTR
7=016 TTR
8=400 GS LS
9= MERMESS 4.0
10= 408 MLX (second attempt, see below - was 100D in 1st attempt)
11= 100D

I have tried two different 100D, 100DCD boards to no avail. I have done board renumber, etc.

On the 1st attempt, the 100DCD lit up all three colors and most of the programming was lost for the entire system. It went to key mode, had random operators and fragments of the original configuration. I pulled the board, did a restore and all was well.

On the 2nd attempt, I added the 408MLX to put a little space between the MerMess and T1 as they are both -5V cards and perhaps affect each other. I modified system programming changes as needed, still no luck.

While in this second attempt, maintenance on the T1 board showed "initialization" and a print of system setup showed several boards, almost all, were "wrong board type".

Once again, i pulled the 408 mlx and 100 DCD, restored, and all was well.

My questions are

1. Since it started as a Legend 7 originally and migrated to Magix via the processor upgrade and MerMess install a few years ago, should I dump the 400 GS LS? It technically isn't supported on magix and perhaps it is messing all subsequent trunk related modules beyond it up.

2. They also had several phantom extension numbers for mailboxes created via "boards" in slots 10-14 as 008 MLX boards with membership in group 30. I originally changed one of the virtual 008 MLX boards to a 100D and did a board renumber. The print all looked good with the phantom 008MLX boards in 11-14 and the extension numbers even bumped up to align. Still I wonder if this whole phantom board thing didn't clear completely.

It is weird that boards in the low slot numbers would show "wrong board" in the print after these failed attempts.

Any thoughts appreciated.

 

[SYQUEST]

Would you please post the print showing the wrong modules? When this happened did you try a board renumber?
What power supplies are in place in the carriers?
Did you check voltages with the 100D or 100DCD modules?

The phantom/adjunct stations should have no effect on your changes. For testing purposes you could drop the 400LSGS and MERMESS modules and put the 100D in slot 8, then do a board renumber. There may be some conflict or defect causing this condition, or even the backplane.

Try some of the suggestions and report back.

....JIM....

 

[exsmogger]

My first thought was the power supply in the expansion since this is an older Legend backplane. I'd recommend upgrading the power supply to a 391C1 if it isn't already.

Brian Cox
Oregon Phone

http://www.OregonPhone.com

 

[phonesaz]

Ditto on the 391C.

If I were in this position I would make sure there is a 391C power supply in both cabinets; if so and the problem continues, I would do a back up to my PC, a frigid start on the processor, and restore from the backup. Frigid starts can make a big difference in clearing up software issues. Keep us posted..

 

[HiDesert]

Acquire a same vintage processor, backplane w/ power supply,
load the current translations into this card and make your
changes...without cards in the system, is it easier to swap
the location of the cards...then install this newly programmed
processor, clear the alarms and see if your system works
properly. It saves the original programming and cuts down on
the time the system is down...

 

[mikedphones]

Thanks for the adice, I have been off the project for a couple of weeks, but it's back on again this week. I will answer some of your questions: 391C in both, yes, Can't easily get a print at the moment, but its the normal "wrong board" message next to most of the mods. A system renumber may fix it, but last time it was under peak business and I really didn't have time so i reverted to the restore with the T1 mod out. I am going to drop the 400GS for sure, I may put the T1 in place of the mermess and bump it one to the right just for a change of environment. I may try frogging the two 391C's anyway. We did lose a couple of 016TTRs last year after a lightning hit, perhaps there is something not right in the 391C that the DS1 boards look for that the others don't. I will try the backup anbd frigid too. i will keep you posted. thanks everyone!

 

[exsmogger]

Sounds like a board renumber would help. The problem is when station mods are removed it changes the numbering for any station mods to the right of the removed mods and can cause all sorts of grief and reprogramming.

Brian Cox
Oregon Phone

http://www.OregonPhone.com

http://www.linkedin.com/in/briancox1952

 

[mikedphones]

Here's the update. Before touching anything, I did a board renumber which erased all the phantom boards. I then pulled the 400 GSLS and replaced it with the 100DCD. The only care beyond it at this point is the Mermess. I did another board renumber.

So far so good. Built the PRI, turned it up, all is well. I then put a 408MLX in the next empty slot to the MerMess and it crapped out again. Print setup showed 408MLX boards all the way up to slot 17! All reading NO FIRMWARE. Pulled the 408, did a board renumber, all went back to normal.

I still need the 50 or so phantom mailboxes, so at this point, I just went into boards and added 3 016MLX cards giving me plenty on both the real and MFM side. The original extension numbers came back when I added these phantom boards. Added them to group 30.

All is well, but I wonder if the backplane isn't the issue. If I didn't have that 400GSLS to pull, I would have been forced to use a new slow, which so far has not worked no matter what. I will revisit this problem if they ever expand and need to add a card.

Thanks to all for your help!