Errors when installing CESM1.2.2 on SeaWulf machine of Stony Brook University

[zhujw@mail_iap_ac_cn]

I tried to install CESM1.2.2 on SeaWulf machine of Stony Brook University.The Atmosphere and land were successfully tested; however, there is a  problem when coupled with the ocean. The information is as follows.

First, it shows  ''Error in `/gpfs/scratch/jiawezhu/output/BC5_test/bld/cesm.exe': malloc(): memory corruption: 0x000000001ea83e40 *** ".Then I debug the model, it further shows **************************************************************************************************forrtl: severe (408): fort: (2): Subscript #3 of the array HUM has value 2 which is greater than the upper bound of 1
Image              PC                Routine            Line        Source
cesm.exe           00000000075D6676  Unknown               Unknown  Unknown
cesm.exe           00000000059BE8E5  overflows_mp_ovf_        5612  overflows.F90
cesm.exe           00000000058BC9D4  overflows_mp_init        1449  overflows.F90
cesm.exe           00000000062C255A  initial_mp_pop_in         345  initial.F90
cesm.exe           0000000005CBD2E0  pop_initmod_mp_po         102  POP_InitMod.F90
cesm.exe           0000000005816394  ocn_comp_mct_mp_o         261  ocn_comp_mct.F90
cesm.exe           0000000000433D4E  ccsm_comp_mod_mp_        1130  ccsm_comp_mod.F90
cesm.exe           00000000004E145F  MAIN__                     90  ccsm_driver.F90
cesm.exe           0000000000412C1E  Unknown               Unknown  Unknown
libc-2.17.so       00002AAAAFF6FB35  __libc_start_main     Unknown  Unknown
cesm.exe           0000000000412B29  Unknown               Unknown  Unknown********************************************************************************************************What I set in file Macros is as follows.*********************************************************************************************************#
# Makefile Macros generated from /gpfs/home/jiawezhu/cesm1.2.2/src/scripts/ccsm_utils/Machines/config_compilers.xml using
# COMPILER=intel
# OS=LINUX
# MACH=SeaWulf
#
CPPDEFS+= -DFORTRANUNDERSCORE -DNO_R16 -DLINUX -DCPRINTEL

SLIBS+=-L/cm/shared/apps/intel/compilers_and_libraries/2017.0.098/mpi/intel64/lib
             -L/cm/shared/apps/intel/compilers_and_libraries/2017.0.098/compiler/lib   
             -L/cm/shared/apps/hdf5_18/1.8.17/lib   
             -L/gpfs/home/jiawezhu/soft/netcdf/lib -lnetcdf -lnetcdff

CFLAGS:= -O0 -fp-model precise

CONFIG_ARGS:=

CXX_LDFLAGS:= -cxxlib

CXX_LINKER:=FORTRAN

ESMF_LIBDIR:=

FC_AUTO_R8:= -r8

FFLAGS:= -mcmodel=medium -no-opt-dynamic-align  -fp-model source -convert big_endian -assume byterecl -ftz -traceback -assume realloc_lhs

FFLAGS_NOOPT:= -O0

FIXEDFLAGS:= -fixed -132

FREEFLAGS:= -freeMPICC:= mpiicc

MPICXX:= mpicxx

MPIFC:= mpiifort

MPI_LIB_NAME:=

MPI_PATH:=/cm/shared/apps/intel/compilers_and_libraries/2017.0.098/mpi/intel64/

NETCDF_PATH:=/gpfs/home/jiawezhu/soft/netcdf

PNETCDF_PATH:=

SCC:= icc

SCXX:= icpc

SFC:= ifort

SUPPORTS_CXX:=TRUE

ifeq ($(compile_threaded), true)
   LDFLAGS += -openmp
   FFLAGS += -openmp
   CFLAGS += -openmp
endififeq ($(MODEL), pop2)
   CPPDEFS += -D_USE_FLOW_CONTROL
endif

ifeq ($(DEBUG), FALSE)
   FFLAGS += -O0
endif

ifeq ($(DEBUG), TRUE)
   FFLAGS += -O0 -g -check uninit -check bounds -check pointers -fpe0
endif******************************************************************************************I guess the problem may be associated with the memory.

Is it possible to add some options to solve these problems when building the model, or any suggestions?Thank you very much.