[quarkz]

Hi,

I tried to run a turbulent case with openfoam. On one cluster, it diverges at the 3rd time step. On another, it completed w/o any problem. I have shown the output below. Is there any reason why? They're using the same config file with the same no. of processors.

Please advice.

cluster1:

Mean and max Courant Numbers = 0.0998 1.067
Flow time scale min/max = 4.687e-07, 2.869e-06
Time = 3e-06

diagonal: Solving for rho, Initial residual = 0, Final residual = 0, No Iterations 0
diagonal: Solving for rhoUx, Initial residual = 0, Final residual = 0, No Iterations 0
diagonal: Solving for rhoUy, Initial residual = 0, Final residual = 0, No Iterations 0
diagonal: Solving for rhoUz, Initial residual = 0, Final residual = 0, No Iterations 0
smoothSolver: Solving for Ux, Initial residual = 0.001212, Final residual = 7.712e-17, No Iterations 2
smoothSolver: Solving for Uy, Initial residual = 0.003422, Final residual = 2.676e-16, No Iterations 2
smoothSolver: Solving for Uz, Initial residual = 0.003535, Final residual = 2.487e-16, No Iterations 2
diagonal: Solving for rhoE, Initial residual = 0, Final residual = 0, No Iterations 0
smoothSolver: Solving for e, Initial residual = 2.144e-05, Final residual = 6.111e-16, No Iterations 2
smoothSolver: Solving for omega, Initial residual = 0.0001931, Final residual = 2.311e-11, No Iterations 2
bounding omega, min: 0 max: 3.156e+05 average: 1449
smoothSolver: Solving for k, Initial residual = 0.1251, Final residual = 4.293e-09, No Iterations 2
bounding k, min: 0 max: 0.9901 average: 0.9261
ExecutionTime = 1.21 s ClockTime = 2 s

Mean and max Courant Numbers = 0.103 1.086
Flow time scale min/max = 4.606e-07, 2.755e-06
Time = 4e-06

cluster2:

Mean and max Courant Numbers = 0.0998 1.067
Flow time scale min/max = 4.687e-07, 2.869e-06
Time = 3e-06

diagonal: Solving for rho, Initial residual = 0, Final residual = 0, No Iterations 0
diagonal: Solving for rhoUx, Initial residual = 0, Final residual = 0, No Iterations 0
diagonal: Solving for rhoUy, Initial residual = 0, Final residual = 0, No Iterations 0
diagonal: Solving for rhoUz, Initial residual = 0, Final residual = 0, No Iterations 0
smoothSolver: Solving for Ux, Initial residual = 0.001212, Final residual = 8.159e-17, No Iterations 2
smoothSolver: Solving for Uy, Initial residual = 0.003422, Final residual = 2.783e-16, No Iterations 2
smoothSolver: Solving for Uz, Initial residual = 0.003535, Final residual = 2.524e-16, No Iterations 2
diagonal: Solving for rhoE, Initial residual = 0, Final residual = 0, No Iterations 0
smoothSolver: Solving for e, Initial residual = 2.144e-05, Final residual = 6.481e-16, No Iterations 2
[21] #0[23] #0 Foam::error:rintStack(Foam::Ostream&)Foam::error :rintStack(Foam::Ostream&)[11] #0 Foam::error:rintStack(Foam::Ostream&)[10] #0 Foam::error:rintStack(Foam::Ostream&)[18] #0 Foam::error:rintStack(Foam::Ostream&)[7] #0 Foam::error:rintStack(Foam::Ostream&) at ??:?
[21] #1 Foam::sigFpe::sigHandler(int) at ??:?
[23] #1 Foam::sigFpe::sigHandler(int) at ??:?
[18] #1 Foam::sigFpe::sigHandler(int) at ??:?
[18] #2 ? at ??:?
[11] #1 Foam::sigFpe::sigHandler(int) at ??:?
[10] #1 Foam::sigFpe::sigHandler(int) at ??:?
[23] #2 ? at ??:?
[21] #2 ? in "/lib/x86_64-linux-gnu/libc.so.6"
[18] #3 Foam::divide(Foam::Field<double>&, Foam::UList<double> const&, Foam::UList<double> const&) in "/lib/x86_64-linux-gnu/libc.so.6"
[21] #3 Foam::divide(Foam::Field<double>&, Foam::UList<double> const&, Foam::UList<double> const&) in "/lib/x86_64-linux-gnu/libc.so.6"
[23] #3 Foam::divide(Foam::Field<double>&, Foam::UList<double> const&, Foam::UList<double> const&) at ??:?
[11] #2 ? at ??:?
[10] #2 ? at ??:?
[18] #4 Foam::tmp<Foam::GeometricField<double, Foam::fvPatchField, Foam::volMesh> > Foam:perator/<Foam::fvPatchField, Foam::volMesh>(Foam::tmp<Foam::GeometricField<doub le, Foam::fvPatchField, Foam::volMesh> > const&, Foam::GeometricField<double, Foam::fvPatchField, Foam::volMesh> const&) in "/lib/x86_64-linux-gnu/libc.so.6"
[11] #3 Foam::divide(Foam::Field<double>&, Foam::UList<double> const&, Foam::UList<double> const&) in "/lib/x86_64-linux-gnu/libc.so.6"
[10] #3 Foam::divide(Foam::Field<double>&, Foam::UList<double> const&, Foam::UList<double> const&) at ??:?
[23] #4 Foam::tmp<Foam::GeometricField<double, Foam::fvPatchField, Foam::volMesh> > Foam:perator/<Foam::fvPatchField, Foam::volMesh>(Foam::tmp<Foam::GeometricField<doub le, Foam::fvPatchField, Foam::volMesh> > const&, Foam::GeometricField<double, Foam::fvPatchField, Foam::volMesh> const&) at ??:?
[21] #4 Foam::tmp<Foam::GeometricField<double, Foam::fvPatchField, Foam::volMesh> > Foam:perator/<Foam::fvPatchField, Foam::volMesh>(Foam::tmp<Foam::GeometricField<doub le, Foam::fvPatchField, Foam::volMesh> > const&, Foam::GeometricField<double, Foam::fvPatchField, Foam::volMesh> const&) at ??:?
[18] #5 Foam::kOmegaSST<Foam::eddyViscosity<Foam::RASModel <Foam::EddyDiffusivity<Foam::ThermalDiffusivity<Fo am::CompressibleTurbulenceModel<Foam::fluidThermo> > > > >, Foam::EddyDiffusivity<Foam::ThermalDiffusivity<Foa m::CompressibleTurbulenceModel<Foam::fluidThermo> > > >::correct() at ??:?
[21] #5 Foam::kOmegaSST<Foam::eddyViscosity<Foam::RASModel <Foam::EddyDiffusivity<Foam::ThermalDiffusivity<Fo am::CompressibleTurbulenceModel<Foam::fluidThermo> > > > >, Foam::EddyDiffusivity<Foam::ThermalDiffusivity<Foa m::CompressibleTurbulenceModel<Foam::fluidThermo> > > >::correct() at ??:?
[23] #5 Foam::kOmegaSST<Foam::eddyViscosity<Foam::RASModel <Foam::EddyDiffusivity<Foam::ThermalDiffusivity<Fo am::CompressibleTurbulenceModel<Foam::fluidThermo> > > > >, Foam::EddyDiffusivity<Foam::ThermalDiffusivity<Foa m::CompressibleTurbulenceModel<Foam::fluidThermo> > > >::correct() at ??:

[quarkz]

I tested with 1 procs, and on another system. I got the same diverged solution after 3 time steps.

So how can I troubleshoot this problem?

Where does it come from and what does the error message says?

Hope someone can help.

Thanks

[quarkz]

Hi,

I finally found the reason for the difference.

In one cluster, the sigfpe has been unset:

unset FOAM_SIGFPE

Hence, the error exception is not captured and stopped.

Is this ok? Or correct

So right now, after running for awhile, at t = 0.005, my residuals are:

diagonal: Solving for rho, Initial residual = 0, Final residual = 0, No Iterations 0
diagonal: Solving for rhoUx, Initial residual = 0, Final residual = 0, No Iterations 0
diagonal: Solving for rhoUy, Initial residual = 0, Final residual = 0, No Iterations 0
diagonal: Solving for rhoUz, Initial residual = 0, Final residual = 0, No Iterations 0
smoothSolver: Solving for Ux, Initial residual = 2.268e-05, Final residual = 1.072e-15, No Iterations 2
smoothSolver: Solving for Uy, Initial residual = 1.904e-06, Final residual = 3.012e-17, No Iterations 2
smoothSolver: Solving for Uz, Initial residual = 1.891e-06, Final residual = 3.097e-17, No Iterations 2
diagonal: Solving for rhoE, Initial residual = 0, Final residual = 0, No Iterations 0
smoothSolver: Solving for e, Initial residual = 7.404e-07, Final residual = 1.87e-16, No Iterations 2
smoothSolver: Solving for omega, Initial residual = 7.563e-06, Final residual = 5.029e-12, No Iterations 2
bounding omega, min: -641.6 max: 1.004e+06 average: 3.899e+04
smoothSolver: Solving for k, Initial residual = 1.925e-05, Final residual = 2.255e-12, No Iterations 2
bounding k, min: -79.06 max: 1.279e+04 average: 339.2

For omega and k, there's the bounding portion there. So are the values reasonable?

Thanks.