Error when running CESM2.3.alpha08a with the SE dynamical core

[tomasz]

Hi everyone,
I am having problems with running CESM2.3.alpha08a with the spectral element dynamical core. I can submit a case but the run terminates after a couple of minutes. Here is an example:

git clone https://github.com/escomp/cesm.git cesm2_3_alpha08a
git checkout cesm2_3_alpha08a

cd cesm2_3_alpha08a/cime/scripts/

./create_newcase --case ~/cases/f.e23.FHIST.ne30_ne30_mg17.test01 --compset FHIST --res ne30_ne30_mg17 --machine fram --project nn1004k --run-unsupported

cd ~/cases/f.e23.FHIST.ne30_ne30_mg17.test01

./case.setup
./case.build
./case.submit

This results in the attached log file (cesm.log.4280297.220329-135201). Also attached are the config_machines.xml, config_compilers.xml and config_batch.xml files I am using. I am able to run CESM2.2.0 with the spectral element dycore and the same config_*.xml files and I am also able to run CESM2.3.alpha08a with a different dynamical core (for example the finite volume dycore with the grid f09_f09_mg17).

My ultimate goal is to run CESM2.3.alpha08a with a user defined variable mesh and in order to do that I need to be able to use the spectral element dycore. Can anyone help me with this? Do you know if this is a machine specific problem or if this is a bug in the model?

All the best,
Tomas.

 

[jedwards]

We are in the process of transitioning to the nuopc driver code as default but not all F cases are ready to do this
yet. Solve this by using the deprecated driver with the flag '--driver mct' on the command line of create_newcase