CfdOF not responding. Running demos and simple model.

[C-monC]

Hi,

On ubuntu 22
I have installed the FreeCad 0.20.2 conda AppImage.
Using installed hisa and cfmesh (not docker but I have tried the docker image).
OpenFoam version 9 from apt install openfoam9.

The error:

It seems like the simulation does complete. After reopening freecad after killing it the solution in paraview returns this error.

I initially tried the demos and assumed they were out of date but I have made a basic rectangular tube example that also results in FreeCad not responding.

Dependancy checker output:

Checking dependencies...
System: Linux
Runtime: Posix
OpenFOAM directory: /opt/openfoam9
Checking dependencies...
System: Linux
Runtime: Posix
OpenFOAM directory: /opt/openfoam9
Paraview executable: /opt/paraviewopenfoam56/bin/paraview
gmsh executable: /usr/bin/gmsh
Completed CFD dependency check
Paraview executable: /opt/paraviewopenfoam56/bin/paraview
gmsh executable: /usr/bin/gmsh
Completed CFD dependency check
Checking dependencies...
System: Linux
Runtime: Posix
OpenFOAM directory: /opt/openfoam9
Paraview executable: /opt/paraviewopenfoam56/bin/paraview
gmsh executable: /usr/bin/gmsh
Completed CFD dependency check

I can confirm that all the demos also freeze FreeCAD.
Memory and CPU usage remain low and I have left the simulation for several hours.

I have attached the FCStd as a zip.
example_freeze.zip

[oliveroxtoby]

This is a puzzler - I am trying with the same AppImage, same OS and same OpenFOAM version and not having any problems. It appears that the residual graph is not appearing on your machine, so perhaps something to do with the plotting. Strange!

[C-monC]

I compiled FreeCad, reinstalled openFoam 9 and I'm using the docker image but I still have the CfdOF freeze right after the simulation completes.

Manually running the simulation works.

cd /tmp/case
simpleFoam # not necessary if I started the sim in freecad

paraview pv.foam

I have also just noticed this error/warning for Qprocess which pops up when the UI freezes.

Do you have any ideas as to where this issue originates from (a specific file/line in this repo)?

[C-monC]

I see it doesn't necessarily finish the simulation (in Cfdof - it finishes if I run simpleFoam). The UAV demo doesn't complete.

Starting time loop

Time = 1

smoothSolver: Solving for Ux, Initial residual = 0.13908732, Final residual = 0.0074376413, No Iterations 1
smoothSolver: Solving for Uy, Initial residual = 0.049523763, Final residual = 0.00094335403, No Iterations 2
smoothSolver: Solving for Uz, Initial residual = 0.063253364, Final residual = 0.00092508066, No Iterations 2
GAMG: Solving for p, Initial residual = 1, Final residual = 0.0053147058, No Iterations 6
GAMG: Solving for p, Initial residual = 0.0033777953, Final residual = 2.3825864e-05, No Iterations 5
GAMG: Solving for p, Initial residual = 0.00064757502, Final residual = 5.5164133e-06, No Iterations 5
GAMG: Solving for p, Initial residual = 0.00014836018, Final residual = 1.3436998e-06, No Iterations 5
GAMG: Solving for p, Initial residual = 4.7266477e-05, Final residual = 4.2853621e-07, No Iterations 5
GAMG: Solving for p, Initial residual = 1.4899016e-05, Final residual = 1.4532247e-07, No Iterations 5
time step continuity errors : sum local = 7.3976029e-07, global = -5.4583983e-07, cumulative = -5.4583983e-07
smoothSolver: Solving for omega, Initial residual = 7.5984468e-05, Final residual = 4.1378296e-06, No Iterations 3
smoothSolver: Solving for k, Initial residual = 1, Final residual = 0.080416255, No Iterations 3
ExecutionTime = 4.27 s ClockTime = 5 s

Time = 2

smoothSolver: Solving for Ux, Initial residual = 0.10961021, Final residual = 0.0049485325, No Iterations 1
smoothSolver: Solving for Uy, Initial residual = 0.10004515, Final residual = 0.0053341389, No Iterations 1
smoothSolver: Solving for Uz, Initial residual = 0.065559269, Final residual = 0.0024366527, No Iterations 1
GAMG: Solving for p, Initial residual = 0.18728523, Final residual = 0.0012793835, No Iterations 5

The last console output before it dies.

[icojb25]

One thing to check, make sure the directory you are storing / running your sim from has write permissions. I have had instances where the default points at a directory which the user cannot write to etc ...

[C-monC]

I changed the output directory to my home directory but FreeCad still hangs after a while.
I'll try on a different ubuntu pc when I get a chance.

[C-monC]

I tried changing the directory to my home directory but Freecad still freezes.
I think I'll just use Cfodf to generate the case folder and manually run the simpleFoam in there for now.

[oliveroxtoby]

Yeah, sorry...no idea at this stage. Really hard to debug this sort of issue without being able to reproduce it. Please let me know if you find out anything further.

[oliveroxtoby]

@C-monC a similar issue was described here on Windows: https://forum.freecad.org/viewtopic.php?t=78096
Workaround was to disable routing python output to the report view. It would be great if you could check whether this avoids your issue as well.