Navier-Stokes demo in parallel fails for n=4

[Felix Ospald]

mpirun -n 4 python demo_navier-stokes.py 
Process 0: Computing tentative velocity
  Process 0: Solving linear system of size 60034 x 60034 (PETSc Krylov solver).
Process 0: Computing pressure correction
  Process 0: Solving linear system of size 7585 x 7585 (PETSc Krylov solver).
[2]PETSC ERROR: --------------------- Error Message ------------------------------------
[2]PETSC ERROR: Argument out of range!
[2]PETSC ERROR: New nonzero at (10,69) caused a malloc!
[2]PETSC ERROR: ------------------------------------------------------------------------
[2]PETSC ERROR: Petsc Release Version 3.4.2, Jul, 02, 2013 
[2]PETSC ERROR: See docs/changes/index.html for recent updates.
[2]PETSC ERROR: See docs/faq.html for hints about trouble shooting.
[2]PETSC ERROR: See docs/index.html for manual pages.
[2]PETSC ERROR: ------------------------------------------------------------------------
[2]PETSC ERROR: Unknown Name on a linux-gnu-cxx-opt named romulus by feo Fri May 23 19:15:00 2014
[2]PETSC ERROR: Libraries linked from /usr/local/FEniCS/1.3.0/lib
[2]PETSC ERROR: Configure run at Mon Jan 13 12:12:39 2014
[2]PETSC ERROR: Configure options --prefix=/usr/local/FEniCS/1.3.0 COPTFLAGS=-O2 --with-debugging=0 --with-shared-libraries=1 --with-clanguage=cxx --with-c-support=1 --download-umfpack=1 --download-hypre=1 --download-mumps=1 --download-scalapack=1 --download-blacs=1 --download-ptscotch=1 --download-scotch=1 --download-metis=1 --download-parmetis=1 --with-ml=1 --with-ml-lib=/usr/local/FEniCS/1.3.0/lib/libml.so --with-ml-include=/usr/local/FEniCS/1.3.0/include/trilinos
[2]PETSC ERROR: ------------------------------------------------------------------------
[2]PETSC ERROR: MatSetValues_MPIAIJ() line 572 in /usr/local/FEniCS/1.3.0/src/petsc-3.4.2/src/mat/impls/aij/mpi/mpiaij.c
[2]PETSC ERROR: MatSetValues() line 1106 in /usr/local/FEniCS/1.3.0/src/petsc-3.4.2/src/mat/interface/matrix.c
[2]PETSC ERROR: MatWrapML_MPIAIJ() line 437 in /usr/local/FEniCS/1.3.0/src/petsc-3.4.2/src/ksp/pc/impls/ml/ml.c
[2]PETSC ERROR: PCSetUp_ML() line 912 in /usr/local/FEniCS/1.3.0/src/petsc-3.4.2/src/ksp/pc/impls/ml/ml.c
[2]PETSC ERROR: PCSetUp() line 890 in /usr/local/FEniCS/1.3.0/src/petsc-3.4.2/src/ksp/pc/interface/precon.c
[2]PETSC ERROR: KSPSetUp() line 278 in /usr/local/FEniCS/1.3.0/src/petsc-3.4.2/src/ksp/ksp/interface/itfunc.c
[2]PETSC ERROR: KSPSolve() line 399 in /usr/local/FEniCS/1.3.0/src/petsc-3.4.2/src/ksp/ksp/interface/itfunc.c
--------------------------------------------------------------------------
mpirun noticed that process rank 2 with PID 8984 on node romulus exited on signal 11 (Segmentation fault).
--------------------------------------------------------------------------

However, it works for n=1,2,3,5,6

[Prof Garth Wells]

Is the demo using BoomerAMG for the Poisson solve? I recall seeing this error periodically in the past with BoomerAMG.

[Felix Ospald]

Thank you. Switching the preconditioner from "amg" to "hypre_amg" worked. With "boomer_amg" it fails.

[Prof Garth Wells]

Works fine for me. Reported error is likely due to old PETSc.