ReconstructPardecomposePar

fabianpk · June 16, 2006, 08:12

I have a case with quite alot of time directories in it. It's run on 4 processors, and I occasionaly reconstruct it to check the results. I've recently gone through the 1.2->1.3 change, but I'm fairly confident this case is run on 1.3.

If I try to decomposePar the case in the directory it's run, I get:

Calculating distribution of cells
Selecting decompositionMethod simple
Foam::error::printStack(Foam:

stream&)
Foam::sigSegv::sigSegvHandler(int)
[0xffffe420]
Foam::primitiveMesh::makeCellCentresAndVols(Foam:: Field<foam::vector<double> > const&, Foam::Field<foam::vector<double> > const&, Foam::Field<foam::vector<double> >&, Foam::Field<double>&) const
Foam::primitiveMesh::calcCellCentresAndVols() const
Foam::primitiveMesh::cellCentres() const
decomposePar [0x810317c]
__gxx_personality_v0
decomposePar [0x8074ab5]
__libc_start_main
__gxx_personality_v0
Segmentation fault (core dumped)

Now come the funny part. I move the time directory I want to decompose, I also move the 0, constant and system to a new directory. Do it again in the new directory, and it works. Could this be a bug?

/Fabian

mattijs · June 16, 2006, 14:55

Is it maybe picking up a new polyMesh/ from somewhere along the way?

June 16, 2006, 08:12	I have a case with quite alot	#1
fabianpk Member Fabian Peng Karrholm Join Date: Mar 2009 Posts: 61 Rep Power: 17	I have a case with quite alot of time directories in it. It's run on 4 processors, and I occasionaly reconstruct it to check the results. I've recently gone through the 1.2->1.3 change, but I'm fairly confident this case is run on 1.3. If I try to decomposePar the case in the directory it's run, I get: Calculating distribution of cells Selecting decompositionMethod simple Foam::error::printStack(Foam:stream&) Foam::sigSegv::sigSegvHandler(int) [0xffffe420] Foam::primitiveMesh::makeCellCentresAndVols(Foam:: Field<foam::vector<double> > const&, Foam::Field<foam::vector<double> > const&, Foam::Field<foam::vector<double> >&, Foam::Field<double>&) const Foam::primitiveMesh::calcCellCentresAndVols() const Foam::primitiveMesh::cellCentres() const decomposePar [0x810317c] __gxx_personality_v0 decomposePar [0x8074ab5] __libc_start_main __gxx_personality_v0 Segmentation fault (core dumped) Now come the funny part. I move the time directory I want to decompose, I also move the 0, constant and system to a new directory. Do it again in the new directory, and it works. Could this be a bug? /Fabian

June 16, 2006, 14:55	Is it maybe picking up a new p	#2
mattijs Senior Member Mattijs Janssens Join Date: Mar 2009 Posts: 1,419 Rep Power: 26	Is it maybe picking up a new polyMesh/ from somewhere along the way?