No Executables, Fatal Errors With module_cu_g3.mod, Etc

All things involving the setup (compiling, NetCDF, etc.) of the WRF Model itself.

No Executables, Fatal Errors With module_cu_g3.mod, Etc

Postby mml373 » Tue Oct 09, 2018 10:41 pm

Hi. I am trying to install version 3.8.1 of WRF on a machine running Ubuntu Linux version 18.04.1 LTS.

gfortran version 7.3.0-27
gcc version 7.3.0-27
netcdf version 4.1.3

Output from log.compile file gives fatal errors similar to the below:

----------
USE module_cu_g3, ONLY : G3INIT
1
Fatal Error: Can't open module file ‘module_cu_g3.mod’ for reading at (1): No such file or directory
compilation terminated.
----------

after "USE module_cu_g3", and also after "USE module_cumulus_driver", "USE module_first_rk_step_part1", and "USE module_physics_init"

I see also that there are lots of warnings for "Nonexistent include directory '/home/blah/Documents/Build_WRF/WRFV3/chem' such as:

----------
f951: Warning: Nonexistent include directory ‘/home/blah/Documents/Build_WRF/WRFV3/chem’ [-Wmissing-include-dirs]
----------


Thanks for help.
mml373
 
Posts: 2
Joined: Tue Oct 09, 2018 9:59 pm

Re: No Executables, Fatal Errors With module_cu_g3.mod, Etc

Postby kwthomas » Wed Oct 10, 2018 3:39 pm

Check your log earlier. Some things didn't compile. Search for "module_cu_g3". My guess is that "cpp" may
have the wrong arguments, so it may have done something wrong. That program is a troublemaker at times.

GFORTRAN is giving harmless warnings about a non-existent empty diriectory. I don't that the Intel compiler
doesn't complain. Just do a "mkdir chem" in your build directory to prevent the complaints.
Kevin W. Thomas
Center for Analysis and Prediction of Storms
University of Oklahoma
kwthomas
 
Posts: 219
Joined: Thu Aug 07, 2008 6:53 pm

Re: No Executables, Fatal Errors With module_cu_g3.mod, Etc

Postby mml373 » Wed Oct 10, 2018 11:21 pm

Thank you. I added the "chem" directory, which solved the problem with the warning. Still have the other problem with the modules.

Looks like there are errors above the line where the module errors begin, such as:
------------
time mpif90 -f90=gfortran -o module_cu_g3.o -c -O2 -ftree-vectorize -funroll-loops -w -ffree-form -ffree-line-length-none -fconvert=big-endian -frecord-marker=4 -I../dyn_em -I../dyn_nmm -I/home/blah/Documents/Build_WRF/WRFV3/external/esmf_time_f90 -I/home/blah/Documents/Build_WRF/WRFV3/main -I/home/blah/Documents/Build_WRF/WRFV3/external/io_netcdf -I/home/blah/Documents/Build_WRF/WRFV3/external/io_int -I/home/blah/Documents/Build_WRF/WRFV3/frame -I/home/blah/Documents/Build_WRF/WRFV3/share -I/home/blah/Documents/Build_WRF/WRFV3/phys -I/home/blah/Documents/Build_WRF/WRFV3/chem -I/home/blah/Documents/Build_WRF/WRFV3/inc -I/home/blah/Documents/Build_WRF/LIBRARIES/netcdf/include module_cu_g3.f90
module_cu_g3.f90:3132:41:

call random_seed (PUT=seed)
1
Error: Size of ‘put’ argument of ‘random_seed’ intrinsic at (1) too small (12/33)
module_cu_g3.f90:3221:41:

call random_seed (PUT=seed)
1
Error: Size of ‘put’ argument of ‘random_seed’ intrinsic at (1) too small (12/33)
module_cu_g3.f90:3236:41:

call random_seed (PUT=seed)
1
Error: Size of ‘put’ argument of ‘random_seed’ intrinsic at (1) too small (12/33)
Command exited with non-zero status 1
0.07user 0.01system 0:00.09elapsed 98%CPU (0avgtext+0avgdata 24528maxresident)k
0inputs+16outputs (0major+7392minor)pagefaults 0swaps
../configure.wrf:334: recipe for target 'module_cu_g3.o' failed
make[3]: [module_cu_g3.o] Error 1 (ignored)


------------

I'll look for more...
mml373
 
Posts: 2
Joined: Tue Oct 09, 2018 9:59 pm

Re: No Executables, Fatal Errors With module_cu_g3.mod, Etc

Postby kwthomas » Thu Oct 11, 2018 3:15 pm

Checking 3.8.1 vs 3.9.1.1 for "module_cu_g3.F" shows a difference that is a bug fix for your problem.

See below. Your line number may vary. The key is the "dimension(12)::seed". Replace that with the new
code.

3125c3125,3126
< integer, dimension (12) :: seed
---
> integer, allocatable :: seed(:)
> integer :: seed_size
3135a3137,3139
> call random_seed(size=seed_size) ! Get size of seed array.
> allocate(seed(1:seed_size)) ! Allocate according to returned
size
>
Kevin W. Thomas
Center for Analysis and Prediction of Storms
University of Oklahoma
kwthomas
 
Posts: 219
Joined: Thu Aug 07, 2008 6:53 pm


Return to Installation

Who is online

Users browsing this forum: No registered users and 9 guests