wrf.exe - Program received signal SIGSEGV: Segmentation fault - invalid memory reference.

[shravan]

The programs geogrid, ungrib, metgrid, and real have run successfully. I have used ERA5 single-level data and ERA5 model-level data (138 levels of data) and produced met_em files (note: instead of regular pressure-level data).
While running wrf.exe as mpirun -np 32 WRF-4.6.0-ARW/run/wrf.exe, I'm facing the following error as in the rsl.error.0009 file.
I have attached the required files below.

Please, can anyone help me in this regard? @William.Hatheway, @kwerner

Many Thanks
-Shravan.

 

[kwerner]

Shravan,
Apologies for the long delay in response while our team tended to time-sensitive obligations. Thank you for your patience.

Are you still experiencing this issue? If so, can you check the met_em* files to make sure you aren't missing any data at any levels? Often when you get a segmentation fault right at the beginning, the issue is specific to the input data. If that doesn't seem to be the case, will you package all of your rsl* files into a single *.tar or zipped file and attach that?

 

[shravan]

Hi @kwerner,

Yes, I’m still facing the same issue. There isn’t any missing data in met_em files. I have tried for 2 domains, also 9 km -> 3 km, as the region is complex (in the Himalayan mountains), even though I am facing the issues. I've attached the rsl zip file (increased debug value in namelist), the met zip file, and the namelist file in Google Drive and am sharing the link here, as the file size is not supported here. Please have a look at it.

Google Drive: Sign-in

Access Google Drive with a Google account (for personal use) or Google Workspace account (for business use).

drive.google.com

-Shravan.

 

[shravan]

Hi @kwerner,

Yes, I’m still facing the same issue. There isn’t any missing data in met_em files. I have tried for 2 domains, also 9 km -> 3 km, as the region is complex (in the Himalayan mountains), even though I am facing the issues. I've attached the rsl zip file (increased debug value in namelist), the met zip file, and the namelist file in Google Drive and am sharing the link here, as the file size is not supported here. Please have a look at it.

Google Drive: Sign-in

Access Google Drive with a Google account (for personal use) or Google Workspace account (for business use).

drive.google.com

-Shravan.

Hi @kwerner,

FYI, I accepted the Google Drive link long back. Could you please confirm whether you received the notification? If so, can you please help me with the problem that I'm facing?

Hope to hear from you soon.

Thanks and Kind regards
-Shravan.

 

[kwerner]

Shravan,
Many apologies for, again, a long delay. We have been swamped with back-to-back workshops this summer, along with staff out of the office, leaving few resources/time to assist with forum inquiries.

Thank you so much for sharing your files with me. I was able to run a test on our NCAR HPC system. I ran it with double the number of processors (64) and it did run longer, but eventually stopped after an hour and 1/2 of simulation time. The error message I get at that point is "WOULD GO OFF TOP." I notice that you have p_top_requested set to 10000, which is lower in the atmosphere than the default value of 5000. Do your data not go higher in the atmosphere? If possible, can you try setting it to 5000 and see if that makes any difference?