[Users] Illegal number of boundary points

[Erik Schnetter]

Maria

The respective check (in line 96) ensures that the number of boundary
points is non-negative, and that it is not larger than the total
number of points on this process. Zero boundary points satisfy both
properties.

Can you check that this condition reads indeed

        if (bnd_points[2*d+f] < 0 || bnd_points[2*d+f] > cctk_lsh[d]) {

on both computers? If so, something is strange on the Ubuntu system.

-erik

On Fri, Mar 9, 2012 at 4:46 PM, Babiuc-Hamilton, Maria
<babiuc at marshall.edu> wrote:
> Hello,
>
> The CarpetReduce thorn does not allow boundary_size = 0, although the Coordbase thorn provides this capabilitiy.
> The error is:
> WARNING level 0 in thorn CarpetReduce processor 0 host babiuc-9
>  (line 96 of /home/babiuc/EinsteinToolkit/Cactus/configs/fourierwave/build/CarpetReduce/mask_coords.c):
>  -> Illegal number of boundary points
> WARNING level 0 in thorn CarpetReduce processor 0 host babiuc-9
>  (line 96 of /home/babiuc/EinsteinToolkit/Cactus/configs/fourierwave/build/CarpetReduce/mask_coords.c):
>  -> Illegal number of boundary points
> cactus_fourierwave: /home/babiuc/EinsteinToolkit/Cactus/configs/fourierwave/build/Carpet/helpers.cc:272: int Carpet::Abort(const cGH*, int): Assertion `0' failed.
> Rank 0 with PID 18884 received signal6
> Writing backtrace to wavetest/backtrace.0.txt
> Aborted
>
> This happens on a Ubuntu. On a MacOS, the problem seems to be overlooked. The parameter file is attached.
>
> thanks,
> Maria
>
>
> _______________________________________________
> Users mailing list
> Users at einsteintoolkit.org
> http://lists.einsteintoolkit.org/mailman/listinfo/users
>



-- 
Erik Schnetter <schnetter at cct.lsu.edu>
http://www.perimeterinstitute.ca/personal/eschnetter/