CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   OpenFOAM Running, Solving & CFD (https://www.cfd-online.com/Forums/openfoam-solving/)
-   -   Problems running icoFoam: sigSegv with 4x mesh resolution (https://www.cfd-online.com/Forums/openfoam-solving/181874-problems-running-icofoam-sigsegv-4x-mesh-resolution.html)

SUBHANKAR December 23, 2016 05:03

Problems running icoFoam: sigSegv with 4x mesh resolution
 
Hi bruno...
Thanks for replying. I will refer to the page you referred. Meanwhile i was solving a 2d flow past cylinder in another system which has openFoam installed before. It gave satisfactory results using icoFoam but while running with finer mesh(original mesh X 4) it gives me this error output:

Code:

/OpenFOAM/OpenFOAM-3.0.1/tutorials/incompressible/icoFoam/kkk$ icoFoam > log
#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::GaussSeidelSmoother::smooth(Foam::word const&, Foam::Field<double>&, Foam::lduMatrix const&, Foam::Field<double> const&, Foam::FieldField<Foam::Field, double> const&, Foam::UPtrList<Foam::lduInterfaceField const> const&, unsigned char, long) at ??:?
#4  Foam::GaussSeidelSmoother::smooth(Foam::Field<double>&, Foam::Field<double> const&, unsigned char, long) const at ??:?
#5  Foam::GAMGSolver::Vcycle(Foam::PtrList<Foam::lduMatrix::smoother> const&, Foam::Field<double>&, Foam::Field<double> const&, Foam::Field<double>&, Foam::Field<double>&, Foam::Field<double>&, Foam::Field<double>&, Foam::Field<double>&, Foam::PtrList<Foam::Field<double> >&, Foam::PtrList<Foam::Field<double> >&, unsigned char) const at ??:?
#6  Foam::GAMGSolver::solve(Foam::Field<double>&, Foam::Field<double> const&, unsigned char) const at ??:?
#7  Foam::fvMatrix<double>::solveSegregated(Foam::dictionary const&) at ??:?
#8  Foam::fvMatrix<double>::solve(Foam::dictionary const&) at ??:?
#9  ? at ??:?
#10  __libc_start_main in "/lib/x86_64-linux-gnu/libc.so.6"
#11  ? at ??:?
Segmentation fault (core dumped)

What could have gone wrong? i have decreased my time step accordingly. Sorry, i know this thread is not meant to solve these sort of problems but i am in a hurry.
regards

wyldckat December 23, 2016 09:40

Greetings Subhankar,
I've moved your post from https://www.cfd-online.com/Forums/op...-failed-2.html to a new thread.

Being in a hurry leads to problems like not checking what information you need to provide us in order to properly help you when any person has the time/opportunity to answer, as explained here: https://www.cfd-online.com/Forums/op...-get-help.html

Based on the information you've provided, the only clues are that:
  • there was a "sigsegv" error, which is a "segmentation fault", usually due to bad memory access;
  • and that you're using GAMG.
Based on these clues, I can only try and guess that the problem might be associated with, so it's possible that you are running the case in parallel and have defined a "coarsest level" value too high for the size of the patches. For example, if you have a patch that has 50 faces inside a single processor domain, and have 100 for the coarsest level, then the algorithm crashes due to a flaw.

But without more details, I cannot say for certain that this is the reason of the crash.

Best regards,
Bruno

SUBHANKAR December 23, 2016 11:04

1 Attachment(s)
Hi bruno,
Thanks again...
i am solving flow past a cylinder for Re=100(2D simulation) and i am not using parallel computation. As i have mentioned it's working fine in lower mesh size like i have simulated it for original mesh size and X2 times the original. All schemes and solvers are kept constant in each case. The problem appears for the case of mesh size X4 times.What other informations do i need to provide? I have attached my case file if anything else you wanted to know...

regards
Subhankar

wyldckat December 27, 2016 12:15

Quick answer/question: I have the case running at the moment, using OpenFOAM 3.0.x. There is one critical detail missing: At which time step does it crash?


edit: I've run for over 9000s (CPU time) and it did not crash when running with OpenFOAM 3.0.x. The output I got at the last iteration before I stopped the run was this:
Code:

Time = 2.67

Courant Number mean: 0.220689 max: 0.589342
DILUPBiCG:  Solving for Ux, Initial residual = 0.000389305, Final residual = 7.37344e-06, No Iterations 2
DILUPBiCG:  Solving for Uy, Initial residual = 0.00108479, Final residual = 8.16693e-07, No Iterations 2
GAMG:  Solving for p, Initial residual = 0.00021379, Final residual = 1.37286e-06, No Iterations 11
GAMG:  Solving for p, Initial residual = 1.38615e-06, Final residual = 8.51044e-07, No Iterations 2
time step continuity errors : sum local = 1.46013e-11, global = 8.67806e-13, cumulative = -5.48055e-09
GAMG:  Solving for p, Initial residual = 0.000123702, Final residual = 8.45245e-07, No Iterations 13
GAMG:  Solving for p, Initial residual = 8.38285e-07, Final residual = 6.92417e-10, No Iterations 33
time step continuity errors : sum local = 1.19764e-14, global = 6.92277e-16, cumulative = -5.48055e-09
ExecutionTime = 9100.57 s  ClockTime = 9112 s

There is a possibility that the problem you're having is due to a bug that was fixed after version 3.0.1 was released.

If you can tell us at which time step the simulation crashes, it would make it a lot easier to try and find the reason for the crash, or least to know how many hours it takes to run the simulation for it to crash.

SUBHANKAR December 29, 2016 23:53

Hi bruno
I had run the simulation twice for the case. In first attempt it crashed at 30sec and next time at 62 sec
Thanks again
regards


All times are GMT -4. The time now is 22:28.