Scheduled Downtime
On Tuesday 24 October 2023 @ 5pm MT the forums will be in read only mode in preparation for the downtime. On Wednesday 25 October 2023 @ 5am MT, this website will be down for maintenance and expected to return online later in the morning.
Normal Operations
The forums are back online with normal operations. If you notice any issues or errors related to the forums, please reach out to help@ucar.edu

Error when running case.build (CESM2.2)

tckm@whu_edu_cn

New Member
There is an error: Deferred-length character component 'tracer_name' at (1) is not yet supported, when running case.build.

/project/sygu/tckm/cesm2.2_beta03/components/clm/src/biogeophys/WaterInfoTracerType.F90:20.49:

character(len=:), allocatable :: tracer_name
1
Error: Deferred-length character component 'tracer_name' at (1) is not yet supported
/project/sygu/tckm/cesm2.2_beta03/components/clm/src/biogeophys/WaterInfoTracerType.F90:20.49:

character(len=:), allocatable :: tracer_name
1
Error: Deferred-length character component 'tracer_name' at (1) is not yet supported
/project/sygu/tckm/cesm2.2_beta03/components/clm/src/biogeophys/WaterInfoTracerType.F90:20.49:

character(len=:), allocatable :: tracer_name
1
Error: Deferred-length character component 'tracer_name' at (1) is not yet supported
/project/sygu/tckm/cesm2.2_beta03/components/clm/src/biogeophys/WaterInfoTracerType.F90:20.49:

character(len=:), allocatable :: tracer_name
1
Error: Deferred-length character component 'tracer_name' at (1) is not yet supported
gmake: *** [WaterInfoTracerType.o] Error 1
gmake: *** Waiting for unfinished jobs....

ERROR: Command gmake complib -j 6 MODEL=clm COMPLIB=/project/sygu/tckm/cesmdata/cesm2.2.0/tet/bld/gnu/openmpi/nodebug/nothreads/mct/mct/esmf/lib/libclm.a -f /project/sygu/tckm/cesm2.2_beta03/cime/scripts/tet/Tools/Makefile CIME_MODEL=cesm compile_threaded=FALSE USE_KOKKOS=FALSE CASEROOT="/project/sygu/tckm/cesm2.2_beta03/cime/scripts/tet" CASETOOLS="/project/sygu/tckm/cesm2.2_beta03/cime/scripts/tet/Tools" CIMEROOT="/project/sygu/tckm/cesm2.2_beta03/cime" COMP_INTERFACE="mct" COMPILER="gnu" DEBUG="FALSE" EXEROOT="/project/sygu/tckm/cesmdata/cesm2.2.0/tet/bld" INCROOT="/project/sygu/tckm/cesmdata/cesm2.2.0/tet/bld/lib/include" LIBROOT="/project/sygu/tckm/cesmdata/cesm2.2.0/tet/bld/lib" MACH="whuatm" MPILIB="openmpi" NINST_VALUE="c1a1l1i1o1r1g1w1i1e1" OS="LINUX" PIO_VERSION="1" SHAREDLIBROOT="/project/sygu/tckm/cesmdata/cesm2.2.0/tet/bld" SMP_PRESENT="FALSE" USE_ESMF_LIB="TRUE" USE_MOAB="FALSE" CAM_CONFIG_OPTS="-phys cam4 -waccmx -ionosphere wxie -chem waccm_ma -offline_dyn -nlev 145" COMPARE_TO_NUOPC="FALSE" USE_ALBANY="FALSE" USE_PETSC="FALSE" failed with rc=2

Does anyone know how to solve this problem?
Thanks very much for replying.
 

Attachments

  • lnd.bldlog.210512-140708.txt
    130.7 KB · Views: 4

nusbaume

Jesse Nusbaumer
CSEG and Liaisons
Staff member
Hello,

I believe that error is due to an out-of-date version of the GNU (gfortran) compiler being used. Do you know if there is a newer version of the GNU compiler on your machine, or a different compiler altogether that is available? If so then I would recommend trying one of those instead.

Of course if you need any help with specifying a new compiler when building CESM just let us know.

Thanks, and have a great day!

Jesse
 

tckm@whu_edu_cn

New Member
Hello,

I believe that error is due to an out-of-date version of the GNU (gfortran) compiler being used. Do you know if there is a newer version of the GNU compiler on your machine, or a different compiler altogether that is available? If so then I would recommend trying one of those instead.

Of course if you need any help with specifying a new compiler when building CESM just let us know.

Thanks, and have a great day!

Jesse
Thanks for your reply.
I compiled the NetCDF, ESMF, and other prerequisites using GNU Fortran (GCC) 4.8.5. Then I create and build cases of CESM 2.2 loading the gcc9.3.
Does this approach also cause the same error above?
The sincerity anticipates your reply.
 

nusbaume

Jesse Nusbaumer
CSEG and Liaisons
Staff member
Hello,

I believe the use of GCC 4.8.5 is what is causing the problem (as the use of deferred-length characters isn't available until 4.9). Can you try building both the prerequisites and the model with GCC 9.3 instead?

Hope that helps, and good luck with the build!

Jesse
 

tckm@whu_edu_cn

New Member
Hello,

I believe the use of GCC 4.8.5 is what is causing the problem (as the use of deferred-length characters isn't available until 4.9). Can you try building both the prerequisites and the model with GCC 9.3 instead?

Hope that helps, and good luck with the build!

Jesse
Thanks for your reply.
Our supercomputer administrator helps me build the prerequisites with GCC 7.3. Then I create and build a new case with GCC 7.3. But there is a fatal error when building.


/project/sygu/tckm/cesm2.2_beta03/cime/src/share/timing/perf_mod.F90:36.7:

use mpi
1
Fatal Error: File 'mpi.mod' opened at (1) is not a GNU Fortran module file
gmake: *** [perf_mod.o] Error 1
ERROR: /project/sygu/tckm/cesm2.2_beta03/cime/src/share/timing/gptl.c: In function GPTLpr_summary_file :
/project/sygu/tckm/cesm2.2_beta03/cime/src/share/timing/gptl.c:3444:8: warning: cast from pointer to integer of different size [-Wpointer-to-int-cast]
if (((int) comm) == 0)
^
/project/sygu/tckm/cesm2.2_beta03/cime/src/share/timing/perf_mod.F90:36.7:

use mpi
1
Fatal Error: File 'mpi.mod' opened at (1) is not a GNU Fortran module file
gmake: *** [perf_mod.o] Error 1


Do you know what causes this error? And how should I solve this problem?
The sincerity anticipates your reply.
 

Attachments

  • gptl.bldlog.txt
    4.5 KB · Views: 0

nusbaume

Jesse Nusbaumer
CSEG and Liaisons
Staff member
Hello,

Do you know what compiler was used to build the version of MPI you are using? My guess is that the error is because MPI was built with a different compiler version then what is being used to build your model/pre-requisites. In other words, if you built all of the libraries (including MPI) with GCC 7.3, then I would try building the model with GCC 7.3 as well.

Also, I am moving this thread to the "infrastructure" forum, as that forum is watched by our porting experts, and they might have a better sense of what is going on (and thus a better solution) then I do.

Hope that helps, and have a great day!

Jesse
 
Top