BC12 upgrade and many lost signals between LSP and TSP

[slash50]

Hi,

We upgraded our customers 14 LIM system to BC12 over the weekend.

Now we have a LOT of lost signals between LSP and TSP, the LIM even forced itself to restart this morning.

We have replaced all TMU cards in the system in accordance with the upgrade procedure, all LPU's have had the firmware changed as USS suggests.

No other LIM has this problem, only LIM 7.

We have repaced the TMU's, even put te old TMU's back. It makes no difference. We've replaced the LPU and restarted both program units and the LIM, no difference.

I'm thinking it has to be software corruption?

Any ideas?

 

[fcpli]

himdp? (or a part of it)

 

[slash50]

TIME DATE ERROR CODE
11:07:20 27FEB06 H'14
JUMP TO ILLEGALSTATE
LOGICAL PROGRAM ERROR ADDRESS = 0200A17E
BRANCHVALUE = 00000006
ENTER
SWSW A-LEVEL SIGNAL CONAUX (H'001D)
FROM TSP (H'33B) EXE A IN LIM 007
TO LSP (H'054) EXE A IN LIM 007
NUMBER OF DATA BYTES IN SIGNAL = 17
WITH 0 1 2 3 4 5 6 7 8 9
000 00 A3 00 09 73 03 3B 01 A8 01
010 FB 65 01 9D 07 02 6A
STACK ADDRESS = 020024B4
SIGNAL SEND ADDRESS = 02005446

USER STACK DUMP
0 1 2 3 4 5 6 7 8 9 A B C D E F
02 00 47 82

TIME DATE ERROR CODE
11:06:38 27FEB06 H'14
JUMP TO ILLEGALSTATE
LOGICAL PROGRAM ERROR ADDRESS = 0200A17E
BRANCHVALUE = 00000006
ENTER
SWSW A-LEVEL SIGNAL CONAUX (H'001D)
FROM TSP (H'33B) EXE A IN LIM 007
TO LSP (H'054) EXE A IN LIM 007
NUMBER OF DATA BYTES IN SIGNAL = 17
WITH 0 1 2 3 4 5 6 7 8 9
000 00 BA 00 08 70 03 3B 01 A8 01
010 FB 64 01 9D 07 02 6A

USER STACK DUMP
0 1 2 3 4 5 6 7 8 9 A B C D E F
02 00 47 82

TIME DATE ERROR CODE
11:06:20 27FEB06 H'14
JUMP TO ILLEGALSTATE
LOGICAL PROGRAM ERROR ADDRESS = 0200A17E
BRANCHVALUE = 00000006
ENTER
SWSW A-LEVEL SIGNAL CONAUX (H'001D)
FROM TSP (H'33B) EXE A IN LIM 007
TO LSP (H'054) EXE A IN LIM 007
NUMBER OF DATA BYTES IN SIGNAL = 17
WITH 0 1 2 3 4 5 6 7 8 9
000 00 B9 00 07 6D 03 3B 01 A8 01
010 FB 64 01 9D 07 02 6A
STACK ADDRESS = 020024B4
SIGNAL SEND ADDRESS = 02005446

USER STACK DUMP
0 1 2 3 4 5 6 7 8 9 A B C D E F
02 00 47 82

TIME DATE ERROR CODE
11:05:38 27FEB06 H'14
JUMP TO ILLEGALSTATE
LOGICAL PROGRAM ERROR ADDRESS = 0200A17E
BRANCHVALUE = 00000006
ENTER
SWSW A-LEVEL SIGNAL CONAUX (H'001D)
FROM TSP (H'33B) EXE A IN LIM 007
TO LSP (H'054) EXE A IN LIM 007
NUMBER OF DATA BYTES IN SIGNAL = 17
WITH 0 1 2 3 4 5 6 7 8 9
000 00 BC 00 06 6A 03 3B 01 A8 01
010 FB 60 01 9D 07 02 6A

USER STACK DUMP
0 1 2 3 4 5 6 7 8 9 A B C D E F
02 00 47 82

TIME DATE ERROR CODE
11:05:20 27FEB06 H'14
JUMP TO ILLEGALSTATE
LOGICAL PROGRAM ERROR ADDRESS = 0200A17E
BRANCHVALUE = 00000006
ENTER
SWSW A-LEVEL SIGNAL CONAUX (H'001D)
FROM TSP (H'33B) EXE A IN LIM 007
TO LSP (H'054) EXE A IN LIM 007
NUMBER OF DATA BYTES IN SIGNAL = 17
WITH 0 1 2 3 4 5 6 7 8 9
000 00 A7 00 05 67 03 3B 01 A8 01
010 FB 60 01 9D 07 02 6A

USER STACK DUMP
0 1 2 3 4 5 6 7 8 9 A B C D E F
02 00 47 82

TIME DATE ERROR CODE
11:04:38 27FEB06 H'14
JUMP TO ILLEGALSTATE
LOGICAL PROGRAM ERROR ADDRESS = 0200A17E
BRANCHVALUE = 00000006
ENTER
SWSW A-LEVEL SIGNAL CONAUX (H'001D)
FROM TSP (H'33B) EXE A IN LIM 007
TO LSP (H'054) EXE A IN LIM 007
NUMBER OF DATA BYTES IN SIGNAL = 17
WITH 0 1 2 3 4 5 6 7 8 9
000 00 B8 00 02 60 03 3B 01 A8 01
010 FB 59 01 9D 07 02 6A
STACK ADDRESS = 020024B4
SIGNAL SEND ADDRESS = 02005446

USER STACK DUMP
0 1 2 3 4 5 6 7 8 9 A B C D E F
02 00 47 82

TIME DATE ERROR CODE
11:04:20 27FEB06 H'14
JUMP TO ILLEGALSTATE
LOGICAL PROGRAM ERROR ADDRESS = 0200A17E
BRANCHVALUE = 00000006
ENTER
SWSW A-LEVEL SIGNAL CONAUX (H'001D)
FROM TSP (H'33B) EXE A IN LIM 007
TO LSP (H'054) EXE A IN LIM 007
NUMBER OF DATA BYTES IN SIGNAL = 17
WITH 0 1 2 3 4 5 6 7 8 9
000 00 BB 00 01 5D 03 3B 01 A8 01
010 FB 59 01 9D 07 02 6A

USER STACK DUMP
0 1 2 3 4 5 6 7 8 9 A B C D E F
02 00 47 82

END

 

[mdman]

restart LSP in this lim

 

[slash50]

I've restarted LSP and TSP already, even restarted the LIM, even done a RFEXI. It's still wobbly, fault escalated to Ericsson now.

Still, any other ideas?

 

[fcpli]

slash:

CONAUX, very "nice". This fault has nothing to do with LSP or TSP, the sequence is just stopping here. What I mean is that the problem occurs earlier, e.g. RVA, trun lines etc. Now, try to find out the traffic case you are meeting this kind of fault. Usually this is only one case you have, meaning e.g. RVA, CIL or something like this. Or, if you can't see the case, try to trace TSP (the sending unit) and figure out which traffic case causes the history. This kind of fault usually depends of a faulty patch (or even code). A good thing is to check which functionalities you have in LIM7 different from others, maybe this is a way to find the fault? But anyway, the fault is NOT in TSP or LSP.

Good luck

 

[slash50]

Ace, thanks mate. We were chasing our tails a bit there. I've an idea it might be to do with older style GJUL cardsand new GJUG cards being incompatible???

 

[dstrang1]

TechnicalUser) 27 Feb 06 15:39 asks:
>Ace, thanks mate. We were chasing our tails a bit there. >I've an idea it might be to do with older style GJUL >cardsand new GJUG cards being incompatible???

This is occuring within the same LIM -- L & G boards are not relevant in this case. Are the LSU & DSU boards in that LIM at the correct rev level for BC12?

I noticed in your HIMDP that the signals are occuring at specific intervals and involve the same addresses.

Have fun,

Dave Strang

 

[cfrei]

Maybe problem is with analog extension. In all these signals is MULTNO of main device 1A8 = 7-1-50-8. Program unit of main device is 19D = ELP6.
We had a lot of problems with quality of line and this signal was in HIMDP too. Usually there was HIGH CALL RATE in ALLOP, but not always.

 

[PCLOP]

to cfrei:
how dp you know that 19D = ELP6? Do you have the other hex values for units?

PCLOP

 

[whosrdaddy]

pclop,

do CNPUP; for that

-----------------------------------------------------
What You See Is What You Get
Never underestimate tha powah of tha google!

 

[slash50]

Problem is fixed, a patch was loaded for ELP6 and now the fault has cleared.

Woooooooohhhoooooooooooooo!! No more lost signals ;o)

Thanks for your efforts everyone )

 

[himdp]

Hey slash50,

I am glad your problem is solved. Can you tell us what the patch number was and do you know why it was not included in the SP release of BC12 you upgraded to?

 

[slash50]

The patch that was activated was AN06006 in ELP6 if that helps. I don't know why it isn't included in the releases up to SP9??

Something to note, it only affected the old "Grey" lim cabinets...

 

[himdp]

Patch AN06006 is actually S110206A and it is too recent to have been included in SP9. The fault has nothing to do with the cabinets, it is specifically when a Voice Mail extension (ICAT D3=1) dials a least cost routing code, >CONAUX> will be sent twice, the second ending in the trash.