[Users] Some quasion about parameter file

[Ian Hinder]

On 13 Mar 2017, at 04:55, 彭兆宏 <alanpeng1995 at gmail.com> wrote:

> 
> 
> Dear Ian,
> 
> Sorry for the delay in replying your letter. The following attachment is my parameter file. I only changed the 
> 
> ReflectionSymmetry::reflection_z = "no", 
> CarpetRegrid2::symmetry_rotating180 = "no", 
> CoordBase::xmin = -120, 
> CoordBase::zmin = -120,
> TwoPunctures::par_S_plus  [2] = 0.10000
> TwoPunctures::par_S_minus[0] = 0.10000
> 
> Which parameters need to be changed when I used this setup?

Hi Alan,

When I run your parameter file, it gives the error

ERROR from host c579.vc-a process 0
  while executing schedule bin BoundaryConditions, routine RotatingSymmetry180::Rot180_ApplyBC
  in thorn RotatingSymmetry180, file /home/ianhin/Cactus/EinsteinToolkitGit/arrangements/CactusNumerical/RotatingSymmetry180/src/rotatingsymmetry180.c:355:
  -> The group "ML_BSSN::ML_LOG_CONFAC" has in the x-direction 63 symmetry points (grid points outside of the symmetry boundary).  This is not equal to the number of ghost zones, which is 3.  The number of symmetry points must be equal to the number of ghost zones.

The problem is that you are running with the RotatingSymmetry180 thorn active, and it is trying to apply this symmetry.  RotatingSymmetry180 behaves differently to ReflectionSymmetry.  ReflectionSymmetry has parameters to enable and disable the symmetry in the different directions, whereas RotatingSymmetry180 only supports a single symmetry direction (a rotation of 180 degrees about the z axis), and so does not have any similar parameters.  Activating the thorn is sufficient to activate the symmetry.  Since you have 

CoordBase::xmin =-120.00

it thinks that all the points between -120 and 0 are symmetry points, which is not correct, so it gives the error.

The fix is to remove RotatingSymmetry180 from ActiveThorns on line 34.  With this change, the parameter file runs for me.

PS: May I give you some advice?  When asking for help to solve an unknown problem such as this, the speed and quality of the help you get back is directly related to how much information you provide.  For example, if you had sent the parameter file and the error message that you got with your original email, then someone familiar with the code would have seen the error message and realised that the problem was related to RotatingSymmetry180.  Since you were trying to run without that symmetry, they would have then told you to make sure it was not active.  They probably would not even have had to try to reproduce the problem.  I think you would have got the answer much more quickly that way.

-- 
Ian Hinder
http://members.aei.mpg.de/ianhin

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.einsteintoolkit.org/pipermail/users/attachments/20170313/73dd499a/attachment.html