agustin@unex_es
Member
This post is similar to a previous one and is a continuation of a post in the CESM General called "too cold to be true". In that post I showed the result of a simulation made with cesm2_1_3 that in the first year of simulation the temperature near the Equator drops to 260K (see the figure).
Following the advice of Erik Kluzek I have made some test. I have begun with the regression test suit. All test in the suit have finish OK except the test_run_restart test with the following message:
FAIL: test_run_restart (__main__.T_TestRunRestart)
----------------------------------------------------------------------
Traceback (most recent call last):
File "./scripts_regression_tests.py", line 1307, in test_run_restart
self._create_test(["NODEFAIL_P1.f09_g16.X"], test_id=self._baseline_name)
File "./scripts_regression_tests.py", line 954, in _create_test
self._wait_for_tests(test_id, expect_works=(not pre_run_errors and not run_errors))
File "./scripts_regression_tests.py", line 963, in _wait_for_tests
from_dir=self._testroot, expected_stat=expected_stat)
File "./scripts_regression_tests.py", line 67, in run_cmd_assert_result
test_obj.assertEqual(stat, expected_stat, msg=msg)
AssertionError:
COMMAND: /lustre/lusitania_homes/UniversidadDeExtremadura/joseagustin.garcia/Documents/CESM/cesm2_1_3/cime/scripts/Tools/wait_for_tests *fake_testing_only_20220209_200621/TestStatus
FROM_DIR: /lusitania_homes/UniversidadDeExtremadura/joseagustin.garcia/Documents/CESM/prueba2/scripts_regression_test.20220209_194502
SHOULD HAVE WORKED, INSTEAD GOT STAT 100
OUTPUT: Test 'NODEFAIL_P1.f09_g16.X.login1_intel' finished with status 'FAIL'
Path: NODEFAIL_P1.f09_g16.X.login1_intel.fake_testing_only_20220209_200621/TestStatus
Analyzing the file TestStatus.log I get the following message:
' failed
See log file for details: /lusitania_homes/UniversidadDeExtremadura/joseagustin.garcia/Documents/CESM/prueba2/scripts_regression_test.20220209_194502/NODEFAIL_P1.f09_g16.X.login1_intel.fake_testing_only_20220209_200621/run/cesm.log.479971.220209-200953
the content of the cesm.log.479971.220209-200953 is
JGF FAKE NODE FAIL
JGF FAKE NODE FAIL
JGF FAKE NODE FAIL
JGF FAKE NODE FAIL
I have also run the create_test with --xml-category prealpha and --xml-machine cheyenne with a lots of FAILS. Find attached the output log.
I don't understand the fact that even in the first year of simulation the model goes so wong. I have ported to my computer in Spain CESM1.2 whith very goods results.
Thanks por your help and patience.
Agustin
Following the advice of Erik Kluzek I have made some test. I have begun with the regression test suit. All test in the suit have finish OK except the test_run_restart test with the following message:
FAIL: test_run_restart (__main__.T_TestRunRestart)
----------------------------------------------------------------------
Traceback (most recent call last):
File "./scripts_regression_tests.py", line 1307, in test_run_restart
self._create_test(["NODEFAIL_P1.f09_g16.X"], test_id=self._baseline_name)
File "./scripts_regression_tests.py", line 954, in _create_test
self._wait_for_tests(test_id, expect_works=(not pre_run_errors and not run_errors))
File "./scripts_regression_tests.py", line 963, in _wait_for_tests
from_dir=self._testroot, expected_stat=expected_stat)
File "./scripts_regression_tests.py", line 67, in run_cmd_assert_result
test_obj.assertEqual(stat, expected_stat, msg=msg)
AssertionError:
COMMAND: /lustre/lusitania_homes/UniversidadDeExtremadura/joseagustin.garcia/Documents/CESM/cesm2_1_3/cime/scripts/Tools/wait_for_tests *fake_testing_only_20220209_200621/TestStatus
FROM_DIR: /lusitania_homes/UniversidadDeExtremadura/joseagustin.garcia/Documents/CESM/prueba2/scripts_regression_test.20220209_194502
SHOULD HAVE WORKED, INSTEAD GOT STAT 100
OUTPUT: Test 'NODEFAIL_P1.f09_g16.X.login1_intel' finished with status 'FAIL'
Path: NODEFAIL_P1.f09_g16.X.login1_intel.fake_testing_only_20220209_200621/TestStatus
Analyzing the file TestStatus.log I get the following message:
' failed
See log file for details: /lusitania_homes/UniversidadDeExtremadura/joseagustin.garcia/Documents/CESM/prueba2/scripts_regression_test.20220209_194502/NODEFAIL_P1.f09_g16.X.login1_intel.fake_testing_only_20220209_200621/run/cesm.log.479971.220209-200953
the content of the cesm.log.479971.220209-200953 is
JGF FAKE NODE FAIL
JGF FAKE NODE FAIL
JGF FAKE NODE FAIL
JGF FAKE NODE FAIL
I have also run the create_test with --xml-category prealpha and --xml-machine cheyenne with a lots of FAILS. Find attached the output log.
I don't understand the fact that even in the first year of simulation the model goes so wong. I have ported to my computer in Spain CESM1.2 whith very goods results.
Thanks por your help and patience.
Agustin