Final restarts incorrect for CICE when IAU is on

[WalterKolczynski-NOAA]

What is wrong?

When IAU is on and the COPY_FINAL_RESTARTS flag is set, workflow attempts to copy the incorrect file names for RUN=gfs.

+ forecast_postdet.sh[740]: case ${RUN} in
+ forecast_postdet.sh[752]: local seconds source_file target_file
+ forecast_postdet.sh[753]: echo 'Copying CICE restarts for '\''RUN=gfs'\'' at 2021033000'
Copying CICE restarts for 'RUN=gfs' at 2021033000
++ forecast_postdet.sh[754]: to_seconds 000000
++ forecast_predet.sh[5]: local hhmmss hh mm ss seconds padded_seconds
++ forecast_predet.sh[6]: hhmmss=000000
++ forecast_predet.sh[7]: hh=00
++ forecast_predet.sh[8]: mm=00
++ forecast_predet.sh[9]: ss=00
++ forecast_predet.sh[10]: seconds=0
+++ forecast_predet.sh[11]: printf %05d 0
++ forecast_predet.sh[11]: padded_seconds=00000
++ forecast_predet.sh[12]: echo 00000
+ forecast_postdet.sh[754]: seconds=00000
+ forecast_postdet.sh[755]: source_file=cice_model.res.2021-03-30-00000.nc
+ forecast_postdet.sh[756]: target_file=20210330.000000.cice_model.res.nc
+ forecast_postdet.sh[757]: /bin/cp -p /work/noaa/stmp/mterry/HERCULES/RUNDIRS/C48mx500_3DVarAOWCDA_28912db2/gfs.2021032500/gfsfcst.2021032500/restart/CICE_RESTART/cice_model.res.2021-03-30-00000.nc /work2/noaa/global/CI/HERCULES/3243/RUNTESTS/COMROOT/C48mx500_3DVarAOWCDA_28912db2/gfs.20210325/00//model/ice/restart/20210330.000000.cice_model.res.nc
/bin/cp: cannot stat '/work/noaa/stmp/mterry/HERCULES/RUNDIRS/C48mx500_3DVarAOWCDA_28912db2/gfs.2021032500/gfsfcst.2021032500/restart/CICE_RESTART/cice_model.res.2021-03-30-00000.nc': No such file or directory

15:35:35 CICE_RESTART/>ls
cice_model.res.2021-03-25-32400.nc  cice_model.res.2021-03-26-75600.nc  cice_model.res.2021-03-28-32400.nc  cice_model.res.2021-03-29-75600.nc
cice_model.res.2021-03-25-75600.nc  cice_model.res.2021-03-27-32400.nc  cice_model.res.2021-03-28-75600.nc
cice_model.res.2021-03-26-32400.nc  cice_model.res.2021-03-27-75600.nc  cice_model.res.2021-03-29-32400.nc

What should have happened?

The appropriate CICE restart file should be copied at the end of the forecast.

What machines are impacted?

All or N/A

What global-workflow hash are you using?

3d6c13e

Steps to reproduce

1. Change COPY_FINAL_RESTARTS to "YES" in config.base
2. Run the C48mx500_3DVarAOWCDA test
3. Check forecast log when gfs forecast fails

Additional information

I think this is an issue with the CICE restarts rather than workflow, as there should always be a restart file at the end of the forecast regardless of IAU and the restart output frequency.

Do you have a proposed solution?

No response