OXO resetting eternally with "SUICIDE OF DB" and "PANIC RESE

[fedebonsi]

Hello everybody, we have one serious issue here. We have an OXO with CPU-3M running R510/023.004, mounted as the following diagram.

Code: Select all

+-----------+----------+-----------+
| AMIX4/4/8 |   APA8   |   CPU-3M  |
+-----------+----------+-----------+
|   UAI16   |  UAI8-1  |  SLI16-1  |
+-----------+----------+-----------+

The OXO DB is empty (default), with the right license.

We noticed that it would never finish startup. Testing it, we found that:

- When OXO has no board connected, it starts correctly and keeps on running without incidents.
- When pushing APA8 board into its slot (2 in this case), OXO instantly resets with error:

Code: Select all

DBMS PB =====> FORCE RESET !!!!

==============================================================
=> PANIC RESET in cmvision.c line 4451 cause : RT_SIG : SUICIDE OF DB
==============================================================

- After resetting, it starts correctly and keeps on running without incidents.
- When running, OXO accepts up to two other boards connected and remains stable.
- If we connect any other board, whichever is it, OXO eventually resets (one or two minutes later), with this error:

Code: Select all

==============================================================
=> PANIC RESET in cmvision.c line 4451 cause : CTRL DSP (80,00) : IS KO
==============================================================

- Then, OXO enters an infinite loop resetting with this cause, and never leaves the startup process.

We thought that there was a problem with the DSP0 of the CPU-3M, but when we connected two other different CPUs, the same error occurred again. Even changing the whole cabinet, back plane and the power supply, and the problem still appeared! Therefore, we put other type of boards that we had in our lab, and worked perfectly! Clearly, where I put the boards (together), I put the problem, but what is the problem??

Have you ever seen something like this? I opened an e-SR and I'm working on it with TS (we have webdiag traces and some other standard info), but we still can't find the solution (nor the cause!).

Thank you in advance!

[steg81_DUP]

I would place cards to this spec,


+-----------+----------+-----------+
| UAI16 | UAI8-1 | CPU-3M |
+-----------+----------+-----------+
| AMIX4/4/8 | SLI16-1| APA8 |
+-----------+----------+-----------+

UAI can't go in slots 4 or 5

[fedebonsi]

Steg81, thanks for your reply.

UAI can't go in slots 4 or 5

In fact, they can according to page 40 of Technical Documentation. However, UAI16 does have a location restriction when using a rack L (slots 5,6,7 and 8 ).

The good news are that I finally located the problem! (although I still don't understand very much the cause of it... ).

The restarting issue happened every time the APA8 or SLI16-1 boards were connected to the system. The first board is the responsible one for the "suicide" thing, and the second one, responsible for the "DSP KO" thing. Apparently there was an OBC problem in this case.

If anyone experiences such a weird case, I hope this could be useful in the future to avoid some long headaches.

Thank you all again.

[tot3nkopf]

Maybe there is a power budget problem. Is this the configuration provided by Actis?

[fedebonsi]

Maybe there is a power budget problem. Is this the configuration provided by Actis?

I thought that at first, but Actis allows the configuration, and the OXO is now working well after changing those two defective boards.

[ACE]

I have a problem with a OXO Release 10, please can u help me?