Scheduled Downtime
On Friday 21 April 2023 @ 5pm MT, this website will be down for maintenance and expected to return online the morning of 24 April 2023 at the latest

WRF segmentation fault with high resolution terrain

gv871

New member
I am attempting to use SRTM terrain data in a WRF simulation. I followed the recommendations from this Thread , and WPS completes successfully. All variables in the generated geo_em.d0#.nc files contain the expected values.
The real.exe step also runs without any apparent issues, and the resulting wrfinput_d0# files contain reasonable and physically consistent fields.
However, when running wrf.exe, the model successfully generates wrfout_d01_######## and wrfout_d02_########, but then terminates with a segmentation fault before producing wrfout_d03_########. Domain 03 is the only domain configured to use the SRTM terrain data.
To isolate the issue, I repeated the simulation using the same configuration but without the SRTM dataset. In that case, the model runs to completion without errors. Based on this comparison, I suspect that the segmentation fault is related to the SRTM terrain data or its configuration in domain 03.
I have attached the namelist.input and the rsl.error on the processor that dumps the error trace.
 

Attachments

  • namelist.input
    4.8 KB · Views: 5
  • rsl_error.txt
    9.7 KB · Views: 2
The rsl file only shows "Program received signal SIGSEGV: Segmentation fault - invalid memory reference", which is not helpful to debug what is wrong.

Please recompile WRF in debug mode, i.e., ./clean -a and ./configure -D, tehn rerun this caase. The rsl will tell excatly when and where the model crashes first. This information can provide you helpful hints what is wrong.
 
Top