[inobsobot]

Dear foamers,


I am trying to run a case with interfoam, derived from the waterChannel tutorial. The geometry is different, as the mesh (snappyhexmesh, checkMesh OK) and initial conditions (see image). Boundary conditions are the same as the tutorial, so are controlDict (except the added function #includeFunc residuals), fvScheme and fvSolution. I run openfoam 7 on ubuntu 20.04 64 bit.


The calculation starts and run smoothly then suddenly crashes without blowing up. What I mean is that everything seems fine (residuals, inlet and outlet fluxes, max velocity) when the crash happens. Here are the last three iterations and the error message :

smoothSolver:  Solving for alpha.water, Initial residual = 0.00119021, Final residual = 7.19883e-09, No Iterations 3
Phase-1 volume fraction = 0.242722  Min(alpha.water) = -4.16565e-09  Max(alpha.water) = 1.00001
MULES: Correcting alpha.water
Phase-1 volume fraction = 0.242722  Min(alpha.water) = -0.00076294  Max(alpha.water) = 1.00039
GAMG:  Solving for p_rgh, Initial residual = 0.0168586, Final residual = 0.000120603, No Iterations 2
time step continuity errors : sum local = 0.000809142, global = 1.39209e-06, cumulative = -9.96522e-06
GAMG:  Solving for p_rgh, Initial residual = 0.000345433, Final residual = 3.85913e-09, No Iterations 22
time step continuity errors : sum local = 2.57429e-08, global = -3.81059e-10, cumulative = -9.9656e-06
smoothSolver:  Solving for omega, Initial residual = 0.00707939, Final residual = 0.000143963, No Iterations 1
smoothSolver:  Solving for k, Initial residual = 0.0212891, Final residual = 0.000628071, No Iterations 1
ExecutionTime = 230.59 s  ClockTime = 231 s

surfaceFieldValue inletFlux write:
    sum(inlet) of rhoPhi = -100

surfaceFieldValue outletFlux write:
    sum(outlet) of rhoPhi = 216.906

surfaceFieldValue atmFlux write:
    sum(atm) of rhoPhi = -0.439954

Courant Number mean: 0.173153 max: 6.19933
Interface Courant Number mean: 0.011299 max: 2.24993
deltaT = 0.00732468
Time = 0.495321

smoothSolver:  Solving for alpha.water, Initial residual = 0.00116221, Final residual = 6.20313e-09, No Iterations 3
Phase-1 volume fraction = 0.242708  Min(alpha.water) = -4.09478e-09  Max(alpha.water) = 1.00001
MULES: Correcting alpha.water
Phase-1 volume fraction = 0.242708  Min(alpha.water) = -0.000681019  Max(alpha.water) = 1.00036
GAMG:  Solving for p_rgh, Initial residual = 0.0164629, Final residual = 0.000117117, No Iterations 2
time step continuity errors : sum local = 0.000745443, global = 3.72879e-07, cumulative = -9.59272e-06
GAMG:  Solving for p_rgh, Initial residual = 0.000343821, Final residual = 4.91956e-09, No Iterations 22
time step continuity errors : sum local = 3.11446e-08, global = -3.32987e-09, cumulative = -9.59605e-06
smoothSolver:  Solving for omega, Initial residual = 0.0067972, Final residual = 0.000129168, No Iterations 1
smoothSolver:  Solving for k, Initial residual = 0.0202467, Final residual = 0.000561078, No Iterations 1
ExecutionTime = 234.09 s  ClockTime = 234 s

surfaceFieldValue inletFlux write:
    sum(inlet) of rhoPhi = -100

surfaceFieldValue outletFlux write:
    sum(outlet) of rhoPhi = 216.794

surfaceFieldValue atmFlux write:
    sum(atm) of rhoPhi = -0.439018

Courant Number mean: 0.168407 max: 6.11036
Interface Courant Number mean: 0.0110264 max: 2.17041
deltaT = 0.0071845
Time = 0.502505

smoothSolver:  Solving for alpha.water, Initial residual = 0.00115311, Final residual = 5.75125e-09, No Iterations 3
Phase-1 volume fraction = 0.242695  Min(alpha.water) = -4.85068e-09  Max(alpha.water) = 1.00001
MULES: Correcting alpha.water
Phase-1 volume fraction = 0.242695  Min(alpha.water) = -0.000757292  Max(alpha.water) = 1.00035
GAMG:  Solving for p_rgh, Initial residual = 0.0161444, Final residual = 0.000116495, No Iterations 2
time step continuity errors : sum local = 0.000719951, global = 1.22603e-07, cumulative = -9.47345e-06
GAMG:  Solving for p_rgh, Initial residual = 0.000345299, Final residual = 4.9922e-09, No Iterations 25
time step continuity errors : sum local = 3.06927e-08, global = -5.35778e-09, cumulative = -9.47881e-06
smoothSolver:  Solving for omega, Initial residual = 0.00660857, Final residual = 0.000119694, No Iterations 1
smoothSolver:  Solving for k, Initial residual = 0.0195878, Final residual = 0.000516912, No Iterations 1
ExecutionTime = 237.7 s  ClockTime = 238 s

surfaceFieldValue inletFlux write:
    sum(inlet) of rhoPhi = -100

surfaceFieldValue outletFlux write:
    sum(outlet) of rhoPhi = 216.687

surfaceFieldValue atmFlux write:
    sum(atm) of rhoPhi = -0.438183

Courant Number mean: 0.166024 max: 6.04775
Interface Courant Number mean: 0.0109053 max: 2.14029
deltaT = 0.00712757
Time = 0.509633

smoothSolver:  Solving for alpha.water, Initial residual = 0.00115602, Final residual = 5.62226e-09, No Iterations 3
Phase-1 volume fraction = 0.242682  Min(alpha.water) = -4.83732e-09  Max(alpha.water) = 1.00001
MULES: Correcting alpha.water
Phase-1 volume fraction = 0.242682  Min(alpha.water) = -0.000996971  Max(alpha.water) = 1.00032
GAMG:  Solving for p_rgh, Initial residual = 0.0161208, Final residual = 0.000117379, No Iterations 2
time step continuity errors : sum local = 0.000718305, global = -1.23836e-07, cumulative = -9.60265e-06
#0  Foam::error::printStack(Foam::Ostream&) at ??:?
#1  Foam::sigSegv::sigHandler(int) at ??:?
#2  ? in "/lib/x86_64-linux-gnu/libc.so.6"
#3  Foam::tmp<Foam::GeometricField<Foam::innerProduct<Foam::GeometricField<Foam::Vector<double>, Foam::fvsPatchField, Foam::surfaceMesh>::value_type, Foam::Vector<double> >::type, Foam::fvsPatchField, Foam::surfaceMesh> > Foam::surfaceInterpolationScheme<Foam::Vector<double> >::dotInterpolate<Foam::GeometricField<Foam::Vector<double>, Foam::fvsPatchField, Foam::surfaceMesh> >(Foam::GeometricField<Foam::Vector<double>, Foam::fvsPatchField, Foam::surfaceMesh> const&, Foam::GeometricField<Foam::Vector<double>, Foam::fvPatchField, Foam::volMesh> const&, Foam::tmp<Foam::GeometricField<double, Foam::fvsPatchField, Foam::surfaceMesh> > const&) at ??:?
#4  Foam::fv::correctedSnGrad<double>::fullGradCorrection(Foam::GeometricField<double, Foam::fvPatchField, Foam::volMesh> const&) const at ??:?
#5  Foam::fv::correctedSnGrad<double>::correction(Foam::GeometricField<double, Foam::fvPatchField, Foam::volMesh> const&) const at ??:?
#6  Foam::fv::snGradScheme<double>::snGrad(Foam::GeometricField<double, Foam::fvPatchField, Foam::volMesh> const&) const at ??:?
#7  Foam::tmp<Foam::GeometricField<double, Foam::fvsPatchField, Foam::surfaceMesh> > Foam::fvc::snGrad<double>(Foam::GeometricField<double, Foam::fvPatchField, Foam::volMesh> const&, Foam::word const&) in "/opt/openfoam7/platforms/linux64GccDPInt32Opt/bin/interFoam"
#8  Foam::tmp<Foam::GeometricField<double, Foam::fvsPatchField, Foam::surfaceMesh> > Foam::fvc::snGrad<double>(Foam::GeometricField<double, Foam::fvPatchField, Foam::volMesh> const&) in "/opt/openfoam7/platforms/linux64GccDPInt32Opt/bin/interFoam"
#9  Foam::interfaceProperties::surfaceTensionForce() const at ??:?
#10  ? in "/opt/openfoam7/platforms/linux64GccDPInt32Opt/bin/interFoam"
#11  __libc_start_main in "/lib/x86_64-linux-gnu/libc.so.6"
#12  ? in "/opt/openfoam7/platforms/linux64GccDPInt32Opt/bin/interFoam"
Segmentation fault (core dumped)

The strangest thing to me is that the exact same simulation will crash at different timestep when run twice in a row. I must also mention the simulation is a single process. Running in parallel only seem to make the simulation crash earlier (first 3 timesteps).


Help would be much appreciated. I can provide details of the case but as I mentioned, it is mostly the water channel tutorial.

[HPE]

Hi,

- Is it possible for you to upload the problematic case?
- Have you found any chance to try this case with OF6, or .com versions? May be the issue is just brand new.

[inobsobot]

Hello HPE, thanks for your time.


I can upload the case, it's about 93 MB. Is there a specific way to do it on the forum or a link to a cloud file would be fine ?


I haven't tried to run the case with OF6. I can try but I am not sure how to install OF6 since it's not in the standard openfoam repository.


Sorry, I am new to this forum and not really proficient in linux.

[HPE]

Hi,

- I don't think there is anything wrong with your simulation (OK, speculation so far - but the segmentation error is a good indicator that something is wrong with the code itself).
- No worries, nobody learned Linux from day-1. I remember that I had been looking on how to open the terminal for roughly five days!
- Installation of any OpenFOAM variant is more or less the same procedure, and you can install all OpenFOAM variants with all its versions at the same time. Have you tried this OF6 website? https://openfoam.org/download/6-ubuntu/
- I think the size of your simulation is 90MB due to the existence of polyMesh directory. Just delete it, we can create the mesh by executing `blockMesh` and `snappyHexMesh` if the relevant settings dictionaries exist for them under `system` directory. If you can upload those text files, I can run the case with different versions to see if the case blows up.

[inobsobot]

Here is the interfoam case. There should be everything needed to mesh and run the case.

[inobsobot]

I solved my problem running the interfoam case on another computer with ubuntu 18.04 and openfoam 7. No Sigsegv and I could also run it in parrallel.


I suppose the problem comes from ubuntu 20.04 or the other computer it is installed on. I don't see what could be wrong with the computer since it was a fresh install of ubuntu 20.04 and openfoam 7.



I don't really know what to do with this conclusion. Maybe report it as a bug but I'm not sure how. If it does come from a version problem between ubuntu 20.04 and openfoam 7, it is quite problematic and should be address.

[HPE]

I'm sorry for the delay for help from my side (I'm still busy). And thanks for your conclusions. You can issue a bug ticket in the Foundation bug tracker website, which you can find below among the links.