[petsc-users] DMPlexDistribute error

[Scott Wales]

Hi,

I'm trying to create a distributed unstructured grid in PETSc, and have encountered the following error:

     [0]PETSC ERROR: --------------------- Error Message ------------------------------------
     [0]PETSC ERROR: Invalid argument!
     [0]PETSC ERROR: Wrong type of object: Parameter # 1!
     [0]PETSC ERROR: ------------------------------------------------------------------------
     [0]PETSC ERROR: Petsc Release Version 3.4.4, Mar, 13, 2014
     [0]PETSC ERROR: See docs/changes/index.html for recent updates.
     [0]PETSC ERROR: See docs/faq.html for hints about trouble shooting.
     [0]PETSC ERROR: See docs/index.html for manual pages.
     [0]PETSC ERROR: ------------------------------------------------------------------------
     [0]PETSC ERROR: bin/celllist_square on a arch-linux2-c-debug named raijin5 by saw562 Mon May 19 13:00:31 2014
     [0]PETSC ERROR: Libraries linked from /home/562/saw562/opt/petsc/3.4.4/lib
     [0]PETSC ERROR: Configure run at Fri May 16 14:23:02 2014
     [0]PETSC ERROR: Configure options --with-shared-libraries=1 --prefix=/home/562/saw562/opt/petsc/3.4.4 --with-blas-lapack-lib="-L/apps/intel-ct/12.1.9.293/mkl/lib/intel64 -lmkl_intel_lp64 -lmkl_sequential -lmkl_core -lpthread" --with-mpi-dir=/apps/openmpi/1.6.3
     [0]PETSC ERROR: ------------------------------------------------------------------------
     [0]PETSC ERROR: ISGetIndices() line 372 in src/vec/is/is/interface/index.c
     [0]PETSC ERROR: DMPlexCreatePartitionClosure() line 2637 ihttps://gist.github.com/ScottWales/2758b5ec96573c63e31an src/dm/impls/plex/plex.c
     [0]PETSC ERROR: DMPlexDistribute() line 2810 in src/dm/impls/plex/plex.c

I've created the DMPlex using `DMPlexCreateFromCellList`, added a default section and then called `DMPlexDistribute` to spread the grid points across all of the processors. You can see my test code at https://gist.github.com/ScottWales/2758b5ec96573c63e31a#file-petsc-test-c-L164. Have I missed a step in the grid setup?

Thanks, Scott