[clau90]

I wrote a Fortran code to implement my outlet boundary conditions in CFX and I created a user CEL function. When I run it on Windows, it works perfectly fine. However, when I try to run my simulations on a Linux Cluster it crashes before the simulation starts.

I re-compiled the code on Linux and updated all the folders and library names. The error I get is:

+--------------------------------------------------------------------+
| Initial Conditions Supplied by Fields in the Input Files |
+--------------------------------------------------------------------+

Domain Name : Pores
Absolute Pressure
Courant Number
Dynamic Viscosity
Effective Density at End of Timestep
Pressure
Pressure.Gradient
Real Partition Number
Shear Strain Rate
Shear Strain Rate.Trnavg
Specific Volume
Total Pressure
User Momentum Source.Bforce
Velocity
Velocity.Beta
Velocity.Gradient
Volume Porosity
Slave: 3 Begin USER_INPUT
Slave: 3 End USER_INPUT

Parallel run: Received message from slave
-----------------------------------------
Slave partition : 3
Slave routine : ErrAction
Master location : RCVBUF,MSGTAG=1214
Message label : 001100279
Message follows below - :

+--------------------------------------------------------------------+
| ERROR #001100279 has occurred in subroutine ErrAction. |
| Message: |
| Signal caught: Segmentation violation |
| |
| |
| |
| |
| |
+--------------------------------------------------------------------+

Parallel run: Received message from slave
-----------------------------------------
Slave partition : 3
Slave routine : ErrAction
Master location : RCVBUF,MSGTAG=1214
Message label : 001100279
Message follows below - :

+--------------------------------------------------------------------+
| ERROR #001100279 has occurred in subroutine ErrAction. |
| Message: |
| Stopped in routine FPX: SIG_HANDLER |
| |
| |
| |
| |
| |
+--------------------------------------------------------------------+

Does anyone have any idea of what this might be?

Thanks a lot,
Claudia

[ghorrocks]

Looks like a bug in your fortran. Check the simulation runs OK without your fortran. Then put some debugging statements (outputting key variables) in your fortran to find the source. You are probably dividing by zero somewhere.

[Antanas]

It seems you try to touch memory which is not allowded to touch. You may try to debug your code to find the actual place.

[clau90]

Thank you for your replies. The simulation works well without the Fortran code. It looks like CFX is able to read the code and execute it (it does output the pressure value when I introduce the line write(*,*)'P=',RET(1,1)):

Begin USER_INPUT
P= 11168.30
End USER_INPUT

I put some debugging statements and it looks like the code crashes right at the end of the subroutine at the line:

CRESLT = 'GOOD'

I am quite surprised because exactly the same Fortran code works on Windows. Any ideas of what this might be?

[ghorrocks]

Are you using the recommended fortran compiler?

[klarke]

Hi, i have the same problem. The CFX version is 15.0, the linux intel fortran compiler is test from 2011~2015. Did you solve the problem?