DOI-USGS / COAWST

COAWST modeling system git repository
Other
104 stars 49 forks source link

Compile Error with COWAST Version 3.8 #244

Open wjlyx opened 5 months ago

wjlyx commented 5 months ago

I attempted to use COAWST 3.8, but encountered some issues while running build_coawst.sh (cannot find m_coupling). I have reviewed the answers to the relevant questions, but I am still unsure how to resolve them. (ROMS+SWAN) Thank you for your assistance.

image build_coawst.txt build.txt

jcwarner-usgs commented 5 months ago

at the command prompt, type $MCT_INCDIR and let me know what it says.

wjlyx commented 5 months ago

at the command prompt, type $MCT_INCDIR and let me know what it says.

Nothing, but I've actually installed MCT, ROMS and WRF coupling to work. Should I have to set MCT_INCDIR? image

jcwarner-usgs commented 5 months ago

yes you need to set export MCT_INCDIR=/pathto_the_inc_dir export MCT_LIBDIR=/pathto_the_lib_dir

wjlyx commented 5 months ago

Hello, I'm trying to get COAWSTv3.8 installed on a big cluster, but I'm still running into this issue. I've got MCT_INCDIR and MCT_LIBDIR set up. image

wjlyx commented 5 months ago

Hello, I'm trying to get COAWSTv3.8 installed on a big cluster, but I'm still running into this issue. I've got MCT_INCDIR and MCT_LIBDIR set up. image

It uses an intel compiler

jcwarner-usgs commented 5 months ago

ok. first what do you get from $MCT_INCDIR second, the error "this module file was not generated by any release of the compiler" is suggesting that you are using different compilers during the build. can you post the entire build output and a copy of your build_coawst.sh? it is probably using mpifort and then mpif90 or some mix.

wjlyx commented 5 months ago

ok. first what do you get from $MCT_INCDIR second, the error "this module file was not generated by any release of the compiler" is suggesting that you are using different compilers during the build. can you post the entire build output and a copy of your build_coawst.sh? it is probably using mpifort and then mpif90 or some mix.For this compilation, I used ROMS and SWAN coupling image build.txt

build_coawst.txt

wjlyx commented 5 months ago

The compiler is Intel and FC=mpiifort in ~/.bashrc. I tried to modify ifort to mpiifort in build_coawst, and added Linux-mpiifort.mk under the Compilers folder, but still encountered the same error as mentioned above.

jcwarner-usgs commented 5 months ago

"/bin/sh: line 24: cmake: command not found" You need to have cmake on your system.

wjlyx commented 5 months ago

"/bin/sh: line 24: cmake: command not found" You need to have cmake on your system.

I have successfully compiled it, although I'm not sure which step was the right one. I've placed build_coawst.sh and the modified Linux-mpiifort.mk here, hoping it will help others. Linux-mpiifort.txt build_coawst.txt

jcwarner-usgs commented 5 months ago

glad you got it to work!

wjlyx commented 5 months ago

glad you got it to work!

glad you got it to work!

Hello, I encountered the following issue while running. Could you please advise on how to resolve it? Thank you. I have checked the rst file and did not find the maximum value for speed. ocean_sandy.txt

jcwarner-usgs commented 5 months ago

that looks like the input file. can you post the output? Also, before i even look at the output file, can you look at the ocean rst file? save the output frequently and look at the his files to see where the problem is developing.

wjlyx commented 5 months ago

that looks like the input file. can you post the output? Also, before i even look at the output file, can you look at the ocean rst file? save the output frequently and look at the his files to see where the problem is developing.

I have reviewed the rst file and did not find any peak values. In the new experiment, I have reduced the time step and it is still running. However, in previous experiments, I encountered an issue. I used a coupled ROMS+WRF model to compare the changes before and after the construction of a wind farm. Many abnormal values appeared on the southeast side. Perhaps you may know what caused this?

The time range is 2015-07-01 to 2015-07-05. The time range is from 2015-07-01 to 2015-07-05. This represents the difference in average wind between the wind farm and non-wind farm on July 1, 2015. image namelist-noowf_input.txt namelist_owf_input.txt ocean_sandy_in.txt qd_windturbines.txt

jcwarner-usgs commented 5 months ago

i am not familiar with adding wind farms to WRF. Suggest you do a run with WRF only and see what that looks like, then do a coupled run.