error when run ./case.build in CESM 2.2.0

Y

Yincheng

Yincheng
New Member
Hi Helpers,

I am using school clusters to config and run CESM2. Everything is good until there is an error happened when run ./case.build, here is the output:
Bash: 
Building case in directory /share/home/czhou171/ycliu/models/CESM/cases/X.test
sharedlib_only is False
model_only is False
Setting Environment MKL_PATH=/share/apps/intel/intel_2019u5/compilers_and_libraries/linux/mkl
Setting Environment NETCDF_HOME=/share/apps/netcdf/
Setting Environment NETCDF_PATH=/share/apps/netcdf/
Generating component namelists as part of build
Creating component namelists
  2022-03-01 16:38:32 atm
   Calling /share/home/czhou171/ycliu/models/CESM/2.2.0/my_cesm_sandbox/cime/src/components/xcpl_comps_mct/xatm/cime_config/buildnml
  2022-03-01 16:38:32 lnd
   Calling /share/home/czhou171/ycliu/models/CESM/2.2.0/my_cesm_sandbox/cime/src/components/xcpl_comps_mct/xlnd/cime_config/buildnml
  2022-03-01 16:38:32 ice
   Calling /share/home/czhou171/ycliu/models/CESM/2.2.0/my_cesm_sandbox/cime/src/components/xcpl_comps_mct/xice/cime_config/buildnml
  2022-03-01 16:38:32 ocn
   Calling /share/home/czhou171/ycliu/models/CESM/2.2.0/my_cesm_sandbox/cime/src/components/xcpl_comps_mct/xocn/cime_config/buildnml
  2022-03-01 16:38:32 rof
   Calling /share/home/czhou171/ycliu/models/CESM/2.2.0/my_cesm_sandbox/cime/src/components/xcpl_comps_mct/xrof/cime_config/buildnml
  2022-03-01 16:38:32 glc
   Calling /share/home/czhou171/ycliu/models/CESM/2.2.0/my_cesm_sandbox/cime/src/components/xcpl_comps_mct/xglc/cime_config/buildnml
  2022-03-01 16:38:32 wav
   Calling /share/home/czhou171/ycliu/models/CESM/2.2.0/my_cesm_sandbox/cime/src/components/xcpl_comps_mct/xwav/cime_config/buildnml
  2022-03-01 16:38:32 iac
   Running /share/home/czhou171/ycliu/models/CESM/2.2.0/my_cesm_sandbox/cime/src/components/stub_comps_mct/siac/cime_config/buildnml
  2022-03-01 16:38:32 esp
   Calling /share/home/czhou171/ycliu/models/CESM/2.2.0/my_cesm_sandbox/cime/src/components/stub_comps_mct/sesp/cime_config/buildnml
  2022-03-01 16:38:32 cpl
   Calling /share/home/czhou171/ycliu/models/CESM/2.2.0/my_cesm_sandbox/cime/src/drivers/mct/cime_config/buildnml
Building gptl with output to file /share/home/czhou171/ycliu/models/CESM/scratch/X.test/bld/gptl.bldlog.220301-163832
   Calling /share/home/czhou171/ycliu/models/CESM/2.2.0/my_cesm_sandbox/cime/src/build_scripts/buildlib.gptl
ERROR: /share/home/czhou171/ycliu/models/CESM/2.2.0/my_cesm_sandbox/cime/src/build_scripts/buildlib.gptl FAILED, cat /share/home/czhou171/ycliu/models/CESM/scratch/X.test/bld/gptl.bldlog.20301-163357

I had a look at the log file(i.e., gptl.bldlog.20301-163357, please see it attached in this thread), it seems like a make problem in the path: /share/home/czhou171/ycliu/models/CESM/2.2.0/my_cesm_sandbox/cime/src/share/timing/. But I am not sure. I would appreciate it if anyone tell me what happened and how to fixed it.

PS: Here is my env varibles and some necessary softwave version in Linux service:
Bash: 
# env vars
export USER=/share/home/czhou171/ycliu
export CESM_CASE=/share/home/czhou171/ycliu/models/CESM/cases
export CESM_PATH=/share/home/czhou171/ycliu/models/CESM/2.2.0/my_cesm_sandbox
# module load
module load python/3.6.8
module load intel/2019u5
module load ncview
module load nco/nco-4.6.3
module load hdf/hdf5-1.8.20_intel
module load gcc/gcc-9.3.0
module load netcdf/4.4_intel
export NETCDF_PATH=/share/apps/netcdf/4.4_intel
# No ESMF
# LAPACK and BLAS provided by MKL
module load cmake/3.16.0
# pnetcdf in local
export PATH="/share/home/czhou171/usr/local/pnetcdf_1.8.1/bin:$PATH"   
export LD_LIBRARY_PATH="/share/home/czhou171/usr/local/pnetcdf_1.8.1/lib:$LD_LIBRARY_PATH"
export INCLUDE="/share/home/czhou171/usr/local/pnetcdf_1.8.1/include:$INCLUDE"

Thanks in advanced!
Yincheng
 

Attachments

J

jedwards

CSEG and Liaisons
Staff member
CESM ignores your shell environment and loads its environment as defined in config_machines.xml for your machine. In
this case it cannot find a c compiler. How are you defining your machine?
 
Y

Yincheng

Yincheng
New Member
jedwards said:
CESM ignores your shell environment and loads its environment as defined in config_machines.xml for your machine. In
this case it cannot find a c compiler. How are you defining your machine?
Click to expand...
Thank you for your timely reply, here is my machine config:
XML: 
<machine MACH="HPCC">
    <DESC>
      Example port to centos7 linux system with gcc, netcdf, pnetcdf and mpich
      using modules from http://www.admin-magazine.com/HPC/Articles/Environment-Modules
    </DESC>
    <!-- should be your regex.expression.matching.your.machine, i.e., run hostname: c01n01 -->
    <NODENAME_REGEX>c01n01</NODENAME_REGEX>
    <OS>RedHat</OS>
      <!-- COMPILERS: compilers supported on this machine, comma seperated list, first is default -->
    <COMPILERS>intel,gnu</COMPILERS>
    <!-- MPILIBS: mpilibs supported on this machine, comma seperated list,
        first is default, mpi-serial is assumed and not required in this list-->
    <MPILIBS>impi, openmpi, mpich</MPILIBS>
      <!-- PROJECT: A project or account number used for batch jobs
      This value is used for directory names. If different from
      actual accounting project id, use CHARGE_ACCOUNT
      can be overridden in environment or $HOME/.cime/config -->
    <PROJECT>czhou171</PROJECT>
    <SAVE_TIMING_DIR> </SAVE_TIMING_DIR>
      <!-- CIME_OUTPUT_ROOT: Base directory for case output,
      the case/bld and case/run directories are written below here -->
    <CIME_OUTPUT_ROOT>$USER/models/CESM/scratch</CIME_OUTPUT_ROOT>
    <DIN_LOC_ROOT>$USER/models/CESM/inputdata</DIN_LOC_ROOT>
    <!-- maybe need to modifiy this path -->
    <DIN_LOC_ROOT_CLMFORC>$USER/models/CESM/inputdata/lmwg</DIN_LOC_ROOT_CLMFORC>
    <DOUT_S_ROOT>$USER/models/CESM/archive/$CASE</DOUT_S_ROOT>
    <BASELINE_ROOT>$USER/models/CESM/cesm_baselines</BASELINE_ROOT>
    <CCSM_CPRNC>$USER/models/CESM/tools/cime/tools/cprnc/cprnc</CCSM_CPRNC>
    <GMAKE>make</GMAKE>
    <GMAKE_J>8</GMAKE_J>
      <!-- please read the menu of Linux clusters to decide batch system -->
    <BATCH_SYSTEM>lsf</BATCH_SYSTEM>
    <SUPPORTED_BY>ycliu@smail.nju.edu.cn</SUPPORTED_BY>
    <!-- please read the menu of Linux clusters to decide the maxiuam kernels in recent node -->
    <MAX_TASKS_PER_NODE>12</MAX_TASKS_PER_NODE>
    <MAX_MPITASKS_PER_NODE>12</MAX_MPITASKS_PER_NODE>
    <PROJECT_REQUIRED>FALSE</PROJECT_REQUIRED>
    <mpirun mpilib="impi">
      <executable>bsub</executable>
      <arguments>
          <arg name="ntasks"> -np {{ total_tasks }} </arg>
      </arguments>
    </mpirun>
      <!-- already manually module load -->
    <module_system type="none"/>
    <!-- <module_system type="module" allow_error="true">
      <init_path lang="perl">/usr/share/Modules/init/perl.pm</init_path>
      <init_path lang="python">/usr/share/Modules/init/python.py</init_path>
      <init_path lang="csh">/usr/share/Modules/init/csh</init_path>
      <init_path lang="sh">/usr/share/Modules/init/sh</init_path>
      <cmd_path lang="perl">/usr/bin/modulecmd perl</cmd_path>
      <cmd_path lang="python">/usr/bin/modulecmd python</cmd_path>
      <cmd_path lang="sh">module</cmd_path>
      <cmd_path lang="csh">module</cmd_path>
      <modules> -->
    <!-- <command name="purge"/>
        </modules>
        <modules compiler="gnu">
    <command name="load">compiler/gnu/8.2.0</command>
    <command name="load">mpi/3.3/gcc-8.2.0</command>
    <command name="load">tool/netcdf/4.6.1/gcc-8.1.0</command>
        </modules>
      </module_system> -->
    <environment_variables>
      <!-- <env name="OMP_STACKSIZE">256M</env> -->
      <env name="MKL_PATH">/share/apps/intel/intel_2019u5/compilers_and_libraries/linux/mkl</env>
      <env name="NETCDF_HOME">/share/apps/netcdf/</env>
      <env name="NETCDF_PATH">/share/apps/netcdf/</env>
    </environment_variables>
    <!-- <resource_limits>
      <resource name="RLIMIT_STACK">-1</resource>
    </resource_limits> -->
  </machine>
should I write the shell environment into the config_machines.xml? Thanks again!
 
Y

Yincheng

Yincheng
New Member
jedwards said:
Yes, your case is trying to load
module compiler/gnu/8.2.0 which doesn't exist on your system
You need to follow the porting guide: Case Control System Part 1: Basic Usage — CIME master documentation
Click to expand...
Thank you for your reply, I followed your guidance and the porting websit to modify my config_machines.xml, but it still the same error output, did I use the wrong c compiler? Here is my config_machines.xml:

XML: 
<machine MACH="HPCC">
    <DESC>
      Example port to centos7 linux system with gcc, netcdf, pnetcdf and mpich
      using modules from http://www.admin-magazine.com/HPC/Articles/Environment-Modules
    </DESC>
    <NODENAME_REGEX>c01n01</NODENAME_REGEX>
    <OS>RedHat</OS>
    <COMPILERS>intel,gnu</COMPILERS>
    <MPILIBS>impi, openmpi, mpich</MPILIBS>
    <PROJECT>czhou171</PROJECT>
    <SAVE_TIMING_DIR> </SAVE_TIMING_DIR>
    <CIME_OUTPUT_ROOT>$USER/models/CESM/scratch</CIME_OUTPUT_ROOT>
    <DIN_LOC_ROOT>$USER/models/CESM/inputdata</DIN_LOC_ROOT>
    <DIN_LOC_ROOT_CLMFORC>$USER/models/CESM/inputdata/lmwg</DIN_LOC_ROOT_CLMFORC>
    <DOUT_S_ROOT>$USER/models/CESM/archive/$CASE</DOUT_S_ROOT>
    <BASELINE_ROOT>$USER/models/CESM/cesm_baselines</BASELINE_ROOT>
    <CCSM_CPRNC>$USER/models/CESM/tools/cime/tools/cprnc/cprnc</CCSM_CPRNC>
    <GMAKE>make</GMAKE>
    <GMAKE_J>8</GMAKE_J>
    <BATCH_SYSTEM>lsf</BATCH_SYSTEM>
    <SUPPORTED_BY>ycliu@smail.nju.edu.cn</SUPPORTED_BY>
    <MAX_TASKS_PER_NODE>12</MAX_TASKS_PER_NODE>
    <MAX_MPITASKS_PER_NODE>12</MAX_MPITASKS_PER_NODE>
    <PROJECT_REQUIRED>FALSE</PROJECT_REQUIRED>
    <mpirun mpilib="impi">
      <executable>bsub</executable>
      <arguments>
        <arg name="ntasks"> -np {{ total_tasks }} </arg>
      </arguments>
    </mpirun>
    <!-- <module_system type="none"/> -->
    <module_system type="module" allow_error="true">
      <init_path lang="perl">/usr/share/Modules/init/perl.pm</init_path>
      <init_path lang="python">/usr/share/Modules/init/python.py</init_path>
      <init_path lang="csh">/usr/share/Modules/init/csh</init_path>
      <init_path lang="sh">/usr/share/Modules/init/sh</init_path>
      <cmd_path lang="perl">/usr/bin/modulecmd perl</cmd_path>
      <cmd_path lang="python">/usr/bin/modulecmd python</cmd_path>
      <cmd_path lang="sh">module</cmd_path>
      <cmd_path lang="csh">module</cmd_path>
      <modules compiler="intel">
      <command name="load">intel/2019u5</command>
      <command name="load">gcc/gcc-9.3.0</command>
      <command name="load">netcdf/4.4_intel</command>
      <command name="load">hdf5-1.8.20_intel</command>
      <command name="load">python/3.6.8</command>
      <command name="load">nco/nco-4.6.3</command>
      <command name="load">ncview</command>
      <command name="load">cmake/3.16.0</command>
      </modules>
    </module_system>
    <environment_variables>
      <!-- <env name="OMP_STACKSIZE">256M</env> -->
      <env name="MKL_PATH">/share/apps/intel/intel_2019u5/compilers_and_libraries/linux/mkl</env>
      <env name="NETCDF_HOME">/share/apps/netcdf/</env>
      <env name="NETCDF_PATH">/share/apps/netcdf/</env>
    </environment_variables>
    <!-- <resource_limits>
      <resource name="RLIMIT_STACK">-1</resource>
    </resource_limits> -->
  </machine>
 
You must log in or register to reply here.
Back
Top