What does this error message mean?

[sivjonsson7]

Hej,
I have got this error several times recently, and I wonder what the meaning of it is, so it is easier to find the error in my code:
"Can't use an undefined value as an ARRAY reference at /opt/noarch/psn/psn-5.2.6/PsN_5_2_6/tool.pm line 1475."
Running using NM7.5, parallel two nodes

The runs have usually finalized and there is a psn-1.lst where I can see that the run has finished, but I may be missing something (probably).

BR, Siv

[rikardn]

It happens when generating the PsN results files. Hard to say exactly why it happens. Is there a way to reproduce the issue?

[sivjonsson7]

Thank you.
I will try to think if I can reproduce it; I have not yet seen a clear pattern in the runs where it happens.

[sivjonsson7]

This is stated in the current -1.lst file:
0THE SIMULATION TASK REQUIRES THAT A BETTER FINAL ESTIMATE
BE AVAILABLE IN THE MODEL SPECIFICATION FILE
OR
MULTIPLE SUB_PROBLEMS IS INCOMPATIBLE WITH OTHER INFORMATION
IN THE CONTROL STREAM AND INPUT MSF

So maybe it is something with the MSF (I am running estimation and simulation in the same model file, using MSF)

[sivjonsson7]

One of the problems was solved by re-parameterizing a parameters (=THETA(#)/1000 since it was estimated to a small value; so problem in my case seems to be related to the SIMULATION using the MSF