Brief summary of bug
PLUMBER2 postad run fails when trying to write a CLM restart file.
General bug information
CTSM version you are using: ctsm5.4.044
Does this bug cause significantly incorrect results in the model's science? No
Configurations affected: postad PLUMBER2 sites
Details of bug
I'm mainly filing this so that when people get this error they can search on it and find this. It's not that obvious what the real problem is from this error message that one gets in the cesm log file:
Abort with message NetCDF: NC_UNLIMITED size already in use in file /glade/derecho/scratch/csgteam/temp/spack/derecho/25.10/builds/spack-stage-parallelio-2.6.8-a4tbnuljlolmvtcsjdkhy4fg2zcqzt2m/spack-src/src/clib/pio_nc.c at line 2107
and this traceback:
cesm.exe 000000000067118C hist_restart_ncd 4540 histFileMod.F90
cesm.exe 00000000006E80DD restfile_write 116 restFileMod.F90
cesm.exe 00000000005B3E64 clm_drv 1536 clm_driver.F90
cesm.exe 000000000056D72A modeladvance 913 lnd_comp_nuopc.F90
The actual cause of this is the setting of these namelist variables in user_nl_clm by the run_tower scripts:
hist_mfilt = 20
hist_nhtfrq = -8760
hist_empty_htapes = .true.
This works fine for AD because the CLM namelist tool will automatically set hist_fincl1:
hist_fincl1 = 'TOTECOSYSC', 'TOTECOSYSN', 'TOTSOMC', 'TOTSOMN', 'TOTVEGC', 'TOTVEGN', 'TLAI', 'GPP', 'CPOOL', 'NPP', 'TWS', 'H2OSNO'
but this is not automatically set for post-AD and hence no history fields will be specified.
It seems like this should be related to #3404 but the error reported is different there.
I will fix this in tower_site.py by explicitly specifying hist_fincl1.
Important details of your setup / configuration so we can reproduce the bug
./run_tower --plumber-sites AU-How --run-type postad --xmlchange STOP_OPTION=nyears,STOP_N=1,PROJECT=P93300041
Brief summary of bug
PLUMBER2 postad run fails when trying to write a CLM restart file.
General bug information
CTSM version you are using: ctsm5.4.044
Does this bug cause significantly incorrect results in the model's science? No
Configurations affected: postad PLUMBER2 sites
Details of bug
I'm mainly filing this so that when people get this error they can search on it and find this. It's not that obvious what the real problem is from this error message that one gets in the cesm log file:
Abort with message NetCDF: NC_UNLIMITED size already in use in file /glade/derecho/scratch/csgteam/temp/spack/derecho/25.10/builds/spack-stage-parallelio-2.6.8-a4tbnuljlolmvtcsjdkhy4fg2zcqzt2m/spack-src/src/clib/pio_nc.c at line 2107
and this traceback:
cesm.exe 000000000067118C hist_restart_ncd 4540 histFileMod.F90
cesm.exe 00000000006E80DD restfile_write 116 restFileMod.F90
cesm.exe 00000000005B3E64 clm_drv 1536 clm_driver.F90
cesm.exe 000000000056D72A modeladvance 913 lnd_comp_nuopc.F90
The actual cause of this is the setting of these namelist variables in user_nl_clm by the run_tower scripts:
hist_mfilt = 20
hist_nhtfrq = -8760
hist_empty_htapes = .true.
This works fine for AD because the CLM namelist tool will automatically set hist_fincl1:
hist_fincl1 = 'TOTECOSYSC', 'TOTECOSYSN', 'TOTSOMC', 'TOTSOMN', 'TOTVEGC', 'TOTVEGN', 'TLAI', 'GPP', 'CPOOL', 'NPP', 'TWS', 'H2OSNO'
but this is not automatically set for post-AD and hence no history fields will be specified.
It seems like this should be related to #3404 but the error reported is different there.
I will fix this in tower_site.py by explicitly specifying hist_fincl1.
Important details of your setup / configuration so we can reproduce the bug
./run_tower --plumber-sites AU-How --run-type postad --xmlchange STOP_OPTION=nyears,STOP_N=1,PROJECT=P93300041