Error for executing `singularity run` command: invalid int value: 'florence'!

[FariborzDaneshvar-NOAA]

I ran ./workflow.sh florence 2018 command in the /scripts directory, but it failed after writing interpolated mesh to disk (see below)

2023-07-22:00:02:22,407 INFO [hurricane_mesh.py:61] Write interpolated mesh to disk...

slurmstepd: error: TMPDIR [/lustre/.tmp] is not writeable
slurmstepd: error: Setting TMPDIR to /tmp
+ singularity run --bind /lustre /lustre/singularity_images//prep.sif setup_ensemble --track-file /lustre/hurricanes/florence_2018_b7937136-4a2f-4024-93c4-1313bf30f113/nhc_track/hurricane-track.dat --output-directory /lustre/hurricanes/florence_2018_b7937136-4a2f-4024-93c4-1313bf30f113/setup/ensemble.dir/ --num-perturbations 10 --mesh-directory /lustre/hurricanes/florence_2018_b7937136-4a2f-4024-93c4-1313bf30f113/mesh/ --sample-from-distribution --sample-rule korobov --hours-before-landfall -1 --date-range-file /lustre/hurricanes/florence_2018_b7937136-4a2f-4024-93c4-1313bf30f113/setup/dates.csv --nwm-file /lustre/static_data/nwm/NWM_v2.0_channel_hydrofabric/nwm_v2_0_hydrofabric.gdb --tpxo-dir /lustre/static_data/tpxo florence 2018
usage: setup_ensemble.py [-h] --track-file TRACK_FILE --output-directory 
                                           OUTPUT_DIRECTORY --date-range-file DATE_RANGE_FILE -n
                                           NUM_PERTURBATIONS --tpxo-dir TPXO_DIR --nwm-file
                                           NWM_FILE --mesh-directory MESH_DIRECTORY
                                           [--sample-from-distribution]
                                           [--sample-rule SAMPLE_RULE] [--quadrature]
                                           [--use-wwm]
                                           name year
setup_ensemble.py: error: argument year: invalid int value: 'florence'
ERROR conda.cli.main_run:execute(47): `conda run python -m setup_ensemble --track-file /lustre/hurricanes/florence_2018_b7937136-4a2f-4024-93c4-1313bf30f113/nhc_track/hurricane-track.dat --output-directory /lustre/hurricanes/florence_2018_b7937136-4a2f-4024-93c4-1313bf30f113/setup/ensemble.dir/ --num-perturbations 10 --mesh-directory /lustre/hurricanes/florence_2018_b7937136-4a2f-4024-93c4-1313bf30f113/mesh/ --sample-from-distribution --sample-rule korobov --hours-before-landfall -1 --date-range-file /lustre/hurricanes/florence_2018_b7937136-4a2f-4024-93c4-1313bf30f113/setup/dates.csv --nwm-file /lustre/static_data/nwm/NWM_v2.0_channel_hydrofabric/nwm_v2_0_hydrofabric.gdb --tpxo-dir /lustre/static_data/tpxo florence 2018` failed. (See above for error)

[SorooshMani-NOAA]

Thanks @FariborzDaneshvar-NOAA I'll look into it.

[SorooshMani-NOAA]

@FariborzDaneshvar-NOAA I'll start doing more tests, but this issue should be resolved if you get latest. See https://github.com/oceanmodeling/ondemand-storm-workflow/commit/682460c118c0cf4c96dbd26a4f13aa6cad8bd99c.

During cleaning up I removed the argument from the script, but forgot to remove it from the workflow call

[FariborzDaneshvar-NOAA]

Thanks @SorooshMani-NOAA for the update. I pulled updated branch from the repo and ran it again. It passed this step and now is setting up the model ...

[SorooshMani-NOAA]

This issue seems to be resolved in the latest iteration. Tested on the 39 member ensemble. I just run into RAM issue due to EC2 instance chosen for the virtual HPC