[CurvingFlow]

I'm running a propeller simulation with SU2 on an HPC, eventually getting to run objective functions and adjoints with the solver. I have what seems to be a very simple problem to fix, but I'm getting an error when I tried to debug both my submission script and config file.
My mesh is in the appropriate SU2 format and has the appropriate zones marked in the combined mesh file. (zone_1.cfg is the zone with rotation and zone_2 is the zone with x direction flow.)

I'm think I'm following the steps here:
https://su2foundation.org/wp-content...6/Kattmann.pdf

I don't have a lot of experience running SU2 but and somewhat familiar with the CFD terminology in the configuration.

Here are parts of what I have in the zone configuration section and input/output of the main config file:

% ------------- MULTIZONE INFORMATION----------------------------- ------------%
%
% Enable multizone mode
MULTIZONE= YES
%
% Config list for zone specific options
CONFIG_LIST = (zone_1.cfg, zone_2.cfg)
%
MULTIZONE_SOLVER= BLOCK_JACOBI
%
% Number of outer iterations (Block-Gauss-Seidel)
OUTER_ITER = 1
%
% Number of internal iterations (dual time method)
% Inner iteration must be 1 for multizone computation.
INNER_ITER= 1

% ------------------------- INPUT/OUTPUT INFORMATION --------------------------%
%
% Mesh input file
MESH_FILENAME= mesh_vol007.su2
%
% Mesh input file format (SU2, CGNS)
MESH_FORMAT= SU2

This is the error code that keeps popping up:

Error in "void CConfig::SetnZone()": 
-------------------------------------------------------------------------
Number of CONFIG_LIST must match number of zones in mesh file.
------------------------------ Error Exit -------------------------------

In the directory where I'm running from I have the full mesh file (mesh_vol007.su2, the main config file, and the two other config files for the respective zones. I am using this command to run:

enable_lmod
module load container_env su2/7.1.1
srun --mpi=pmi2 crun SU2_CFD mainConfigFile.cfg

Any help would be appreciated. Thanks in advance.

[TKatt]

Hi,


out of the hip I am guessing you are missing the

NZONE= 2

part in your .su2 mesh. See Slide 7 in the presentation you already linked


best, tobi

[CurvingFlow]

Nope I don't think so. Here's how the mesh file begins when I open it in a text editor.

NZONE = 2
IZONE = 1
%
% Problem dimension
%
NDIME= 3
%
% Inner element connectivity
%
NELEM= 11019177
14  112753 112754 112755 112756 112757    0
14  112754 112753 112758 112759 112757    1
14  112755 112754 112759 112760 112757    2
14  112756 112755 112760 112761 112757    3
14  112753 112756 112761 112758 112757    4
14  112762 112763 112764 112765 112766    5

[CurvingFlow]

While comparing files that have worked previously, I am suspecting that the syntax needs to be:

NZONE= 2
IZONE= 1

and not:

NZONE = 2
IZONE = 1

The addition of a space character before the equal sign may have triggered the error. I have not confirmed this yet, but I think this may be the solution.

Thanks for the help, by the way.

[TKatt]

Yep, I can reproduce the issue on my side... this additional space for NZONE, IZONE or NDIME before the = sign makes the setup fail.


I'll take a look whether that is easily fixable. Until then it should work without this space