Segmentation fault (signal 11)

[fiorela_castillon]

I am running WRF model v3.8.1 with gnu compilers in my PC (Intel® Core™ i7-4790 CPU @ 3.60GHz × 8). Preprocessing (GFS 0.25°) and real.exe seems to be ok, but when I execute (even with ulimit -s unlimited) mpirun -np 8 ./wrf.exe a segmentation fault message appears in few seconds. My namelist.input and namelist.wps are attached.

I would appreciate any comments.

Thanks

 

[kwerner]

Hi,

Can you package all of your rsl* files together in a *.tar file and attach it here so that I can take a look?

As a side note, I notice that your domains are pretty small (d02 is 88x88). We don't really recommend using domains any smaller than about 100x100, or the model isn't capable of resolving reasonable results before the parcels are propagated through to the other side. You may want to consider increasing the size of both domains.

 

[fiorela_castillon]

I have incremented mi domain but the segmentation fault persists. I attach my namelists and all the rsl. in a .tar file. Thank you so much for your help kwerner.

 

[kwerner]

Thank you for sending that. The rsl files did not provide much useful information, unfortunately. There are a few things you can check on:

1) Verify that you do have enough memory and that you are not running out of disk space
2) If there are no problems with 1), you can compile the code using 'configure -d' to see if you can get some extra traceback/debugging statements in your rsl files. This should tell you where in the code the error is occurring. Do not use the debug_level option in the namelist. This only adds extra useless junk to the rsl files.
3) You may want to check your wrfinput* files to see if you see anything erroneous - such as soil moisture
4) Can you try setting damp_opt to 3 and see if it makes any difference?