[Users] problem with par file not finding thorns

[Ian Hinder]

On 8 Aug 2015, at 03:05, David Bradley <kb9qhd at charter.net> wrote:

> Greetings
> 
> I tried to run full-040.par simulation from the examples directory and it failed to find several of the thorns.
> 
> Error: Thorn ADMConstraints not found
> Error: Thorn BSSN_MoL not found
> Error: Thorn LegoExcision not found
> Activation failed - 3 errors in activation sequence
> 
> I am not sure how to proceed

Hi David,

(I think you are talking about full-040.par inside the EinsteinInitialData/IDAxiBrillBH/par directory.)

Many of the thorns in the Einstein Toolkit have example parameter files which use thorns which are themselves not a part of the toolkit.  This means that those other thorns are not currently recommended or supported by the ET project.  This happened because these thorns predated the toolkit, and not all existing thorns were added to it.  In your case, the example you are trying to run uses such thorns, so you won't be able to run it with the Einstein Toolkit.

Given that all of the example parameter files for IDAxiBrillBH rely on thorns which are not in the toolkit, that leaves this thorn as somewhat lacking in examples!  Since axisymmetric brill wave initial data is not used very much these days for science, it's unlikely that anyone has a working example parameter file that you could use.

You could find these thorns, add them to your thornlist, and try to compile them, noting that they have probably not been tested or even compiled in years.  ADMConstraints should work (I think I have used it recently), but I don't know about the other two.

Or you could adapt the example parameter file to use supported thorns.  BSSN_MoL would be replaced by the current BSSN evolution code ML_BSSN (though the parameter names are different), but LegoExcision probably doesn't have a more modern equivalent, so maybe you are stuck there.  Excision as a method is not supported in the ET as far as I know.  You could try evolving with a moving puncture method, but this is possibly breaking new ground, and it might not be a good place to start.

Both of these are a fairly large amount of work, depending on your familiarity with the toolkit, and it's not clear what the benefit would be of running this example parameter file.  If you could be more specific about what you are trying to do, we might be able to advise you of the best way forward.  Just taking the initial data parameters from the examples should work and produce the initial data.  Do you then want to evolve it?

If you want a working parameter file which just creates the initial data, then take a look in the "test" directory.  Both these test parameter files are regularly tested and work properly.

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