[CIG-SHORT] How to determine the value of Up direction

Brad Aagaard baagaard at usgs.gov
Wed Aug 30 09:51:16 PDT 2017


Tu Xiang,

Here is the error I get when I use your mesh:

Fatal error. Calling MPI_Abort() to abort PyLith application.
Traceback (most recent call last):
   File 
"/tools/common/pylith-develop/gcc-5.4.0/lib/python2.7/site-packages/pylith/apps/PetscApplication.py", 
line 74, in onComputeNodes
     self.main(*args, **kwds)
   File 
"/tools/common/pylith-develop/gcc-5.4.0/lib/python2.7/site-packages/pylith/apps/PyLithApp.py", 
line 128, in main
     self.problem.initialize()
   File 
"/tools/common/pylith-develop/gcc-5.4.0/lib/python2.7/site-packages/pylith/problems/TimeDependent.py", 
line 120, in initialize
     self.formulation.initialize(self.dimension, self.normalizer)
   File 
"/tools/common/pylith-develop/gcc-5.4.0/lib/python2.7/site-packages/pylith/problems/Explicit.py", 
line 110, in initialize
     self._initialize(dimension, normalizer)
   File 
"/tools/common/pylith-develop/gcc-5.4.0/lib/python2.7/site-packages/pylith/problems/Formulation.py", 
line 470, in _initialize
     integrator.initialize(totalTime, numTimeSteps, normalizer)
   File 
"/tools/common/pylith-develop/gcc-5.4.0/lib/python2.7/site-packages/pylith/bc/AbsorbingDampers.py", 
line 128, in initialize
     BoundaryCondition.initialize(self, totalTime, numTimeSteps, normalizer)
   File 
"/tools/common/pylith-develop/gcc-5.4.0/lib/python2.7/site-packages/pylith/bc/BoundaryCondition.py", 
line 128, in initialize
     ModuleBoundaryCondition.initialize(self, self.mesh(), self.upDir)
   File 
"/tools/common/pylith-develop/gcc-5.4.0/lib/python2.7/site-packages/pylith/bc/bc.py", 
line 130, in initialize
     def initialize(self, *args): return 
_bc.BoundaryCondition_initialize(self, *args)
RuntimeError: Error computing orientation of cell face. Cannot resolve 
tangential components into unambigious directions.
Up direction (0, 0, 1)  cannot be parallel to the face normal (0, 0, -1).
If the face is horizontal, adjust the up_dir parameter.


If you look at the stack trace you will see that this error is 
associated with an absorbing boundary. Obviously the bottom boundary is 
horizontal, so the up_dir for the bottom bondary needs to be changed to 
(1,0,0) or (0,1,0). Sorry for thinking this was for the fault.

Regards,
Brad



More information about the CIG-SHORT mailing list