Error message with WRF installation with dm+sm option

[llsalazardom]

I have a WRF installation with the dm+sm option, I have tested it with a small test and it works fine but with a bigger test I have problems with the real.exe. After about two minutes running I get the following error message:

Input data is acceptable to use: met_em.d04.2019-06-28_05:00:00.nc
metgrid input_wrf.F first_date_input = 2019-06-28_05:00:00
metgrid input_wrf.F first_date_nml = 2019-06-28_05:00:00
[cn0106:1223641:0:1223641] Caught signal 11 (Segmentation fault: address not mapped to object at address 0x145718184cd0
)
==== backtrace (tid:1223641) ====
0 0x0000000000012d10 __funlockfile() :0
1 0x00000000000a7993 __memmove_avx_unaligned_erms() :0
.
.
.

Can someone please help me with this problem?
I will attach the files

 

[islas]

Given the size of the domain & stack trace you may want to look at Determine MPI Data Types in col_on_comm() & dst_on_comm() to prevent displacements overflow. (Fix for #2156) by benkirk · Pull Request #2157 · wrf-model/WRF
If you are using the latest version of WRF from a git repository, this is in the upcoming v4.7.0 release (GitHub - wrf-model/WRF at release-v4.7.0)

However, if you would rather try out only the small patch you can either use the diff from that PR or follow the instructions from here : cxil_map: write error with real.exe