[CIG-SEISMO] SPECFEM3D: Segmentation fault - xgenerate_databases

[Elliott Sales de Andrade]

Hi Gian,

It is difficult to determine the cause of the problem from just the
addresses. You should build with debugging information so that you can
find out the exact location of the problem. The flags.guess file
contains examples of the flags that need to be given to the compiler to
get that debugging information.

Also, I'd like to point out this line:
cp: cannot stat `/OUTPUT_FILES/output_mesher.txt': No such file or
directory
This seems to indicate that your script is not entirely correct,
possibly an undefined variable.

On Thu, 2013-02-14 at 12:27 -0800, gmatharu at eos.ubc.ca wrote:
> Hi
> 
>  I have been running SPECFEM3D on a couple of different clusters on the
> Westgrid computer consortium. As I'm not sure what details are relevant
> I've included links which explain the builds of either cluster.
> 
> http://www.westgrid.ca/support/quickstart/lattice
> http://www.westgrid.ca/support/quickstart/orcinus
> 
> I'm currently generating a mesh on 256 processors which is 256*128*60
> (X/Y/Z). A few months back on an older build of the software I had no
> issues generating a mesh on the orcinus cluster and running the solver.
> However with an updated package I am encountering segmentation fault
> errors when running xgenerate_databases on both Lattice and Orcinus (using
> ifort,icc and mpif90 on both). My submission scripts seem correct and the
> initial meshing seems to work ok, however I'm at a loss as to what's
> causing this error. Do you have any recommendations on how to resolve this
> issue? If you require further information let me know.
> 
> Best Regards,
> 
> Gian
> 
> Below is a snippet of the error message I receive.
> 
> 
> starting MPI internal mesher on 256 processors
> 
> generating databases...
> cp: cannot stat `/OUTPUT_FILES/output_mesher.txt': No such file or directory
> forrtl: severe (174): SIGSEGV, segmentation fault occurred
> Image              PC                Routine            Line        Source
> libifcore.so.5     00002B8064AF3683  Unknown               Unknown  Unknown
> xgenerate_databas  0000000000473530  Unknown               Unknown  Unknown
> xgenerate_databas  000000000045248E  Unknown               Unknown  Unknown
> xgenerate_databas  0000000000417D08  Unknown               Unknown  Unknown
> xgenerate_databas  0000000000416351  Unknown               Unknown  Unknown
> xgenerate_databas  000000000040D10B  Unknown               Unknown  Unknown
> xgenerate_databas  000000000040BD09  Unknown               Unknown  Unknown
> xgenerate_databas  000000000040BC9C  Unknown               Unknown  Unknown
> libc.so.6          0000003F7FA1D994  Unknown               Unknown  Unknown
> xgenerate_databas  000000000040BBA9  Unknown               Unknown  Unknown
> forrtl: severe (174): SIGSEGV, segmentation fault occurred
> Image              PC                Routine            Line        Source
> libifcore.so.5     00002B2EB0DF5683  Unknown               Unknown  Unknown
> xgenerate_databas  0000000000473530  Unknown               Unknown  Unknown
> xgenerate_databas  000000000045248E  Unknown               Unknown  Unknown
> xgenerate_databas  0000000000417D08  Unknown               Unknown  Unknown
> xgenerate_databas  0000000000416351  Unknown               Unknown  Unknown
> xgenerate_databas  000000000040D10B  Unknown               Unknown  Unknown
> xgenerate_databas  000000000040BD09  Unknown               Unknown  Unknown
> xgenerate_databas  000000000040BC9C  Unknown               Unknown  Unknown
> libc.so.6          000000336A21D994  Unknown               Unknown  Unknown
> xgenerate_databas  000000000040BBA9  Unknown               Unknown  Unknown
> 
> 
> _______________________________________________
> CIG-SEISMO mailing list
> CIG-SEISMO at geodynamics.org
> http://geodynamics.org/cgi-bin/mailman/listinfo/cig-seismo