Hello ZhiChao,

However, I found a lot of logs that likes : “New star location not set (18.4258,0.763088,0) because too far from old (15.0469,1.82812,0).”

I think maybe it shows there is some error.

Could you list some possible unreasonable uses that that can cause this error?

My guess would be that you are not searching for the stars often enough so that they have moved too much in between times they are found.

If parfile and initial_data are necessary, my parfile and the initial_data are here: https://github.com/yanyuechuixue/parfiles

Looking at your parfile the NRTracker and Hydro_Analysis settings seem ok since you are using the defaults which will looks for stars at every step.

You are however using IllinoisGRMHD which comes with its own tracker thorn, VolumeIntegralsGRMHD which was tested with it (by setting its amr_centre__tracks__volintegral_inside_sphere paramters). In particular I am a bit worried about you setting

convert_to_HydroBase::convert_to_HydroBase_every = 8

which would seem (I have not used IGMHD myself) indicate that the HydroBase variables (which Hydro_Analysis uses) are only updated every 8 steps which is too infrequent given that Hydro_Analysis needs this data every single step.

Since you are using both VolumeIntegralsGRMHD and Hydro_Analysis you have two thorns enabled that can track stars.

Yours, Roland