deepmodeling / abacus-develop

An electronic structure package based on either plane wave basis or numerical atomic orbitals.
http://abacus.ustc.edu.cn
GNU Lesser General Public License v3.0
173 stars 134 forks source link

Test: our CICD auto test workflow does not work fully as expected #4335

Open kirk0830 opened 5 months ago

kirk0830 commented 5 months ago

Describe the Testing Issue

see Action: https://github.com/deepmodeling/abacus-develop/actions/runs/9412701654/job/25928061867?pr=4324 The cell_parallel_kpoints_test also failed but in the summary it was like:


99% tests passed, 1 tests failed out of 296

Total Test time (real) = 2292.92 sec

The following tests FAILED:
     78 - cell_klist_test (Subprocess aborted)
Errors while running CTest
Output from these tests are in: /__w/abacus-develop/abacus-develop/build/Testing/Temporary/LastTest.log
Use "--rerun-failed --output-on-failure" to re-run the failed cases verbosely.
gmake: *** [Makefile:71: test] Error 8

see Action: https://github.com/deepmodeling/abacus-develop/actions/runs/10001403608/job/27644946382?pr=4733 At least two unittests exit with code 0 but actually not work as expected:

190/301 Test #190: hontainer_para_test ...............   Passed    2.16 sec
test 191
        Start 191: hcontainer_output_test
191: Test command: /__w/abacus-develop/abacus-develop/build/source/module_hamilt_lcao/module_hcontainer/test/hcontainer_output_test
191: Test timeout computed to be: 1700
191: Running main() from ./googletest/src/gtest_main.cc
191: [==========] Running 1 test from 1 test suite.
191: [----------] Global test environment set-up.
191: [----------] 1 test from OutputHContainerTest
191: [ RUN      ] OutputHContainerTest.Write
191/301 Test #191: hcontainer_output_test ............   Passed    0.67 sec
test 192
        Start 192: hcontainer_readCSR_test
192: Test command: /__w/abacus-develop/abacus-develop/build/source/module_hamilt_lcao/module_hcontainer/test/hcontainer_readCSR_test
192: Test timeout computed to be: 1700
192: Running main() from ./googletest/src/gtest_main.cc
192: [==========] Running 1 test from 1 test suite.
192: [----------] Global test environment set-up.
192: [----------] 1 test from ReadHContainerTest
192: [ RUN      ] ReadHContainerTest.ReadAndOutputHContainer
192:  
192:  !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
192:                          NOTICE                           
192:  !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
192:  
192:  row or col index out of range
192:  CHECK IN FILE : warning.log
192:  
192:  !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
192:                          NOTICE                           
192:  !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
192: TIME STATISTICS
192: -----------------------------------------------------
192:  CLASS_NAME    NAME    TIME/s  CALLS   AVG/s  PER/%  
192: -----------------------------------------------------
192:             total      0.00   1        0.00   -nan   
192: -----------------------------------------------------
192: 
192:  See output information in : 
192/301 Test #192: hcontainer_readCSR_test ...........   Passed    0.74 sec
test 193
        Start 193: deltaspin_parse_json_test
193: Test command: /__w/abacus-develop/abacus-develop/build/source/module_hamilt_lcao/module_deltaspin/test/deltaspin_parse_json_test
193: Test timeout computed to be: 1700

not any [ OK ] appears. It may because the exit(0) used in WARNING_QUIT function.

Additional Context

No response

Task list for Issue attackers (only for developers)

caic99 commented 5 months ago

https://github.com/deepmodeling/abacus-develop/blob/332c875f328f3da8975059d4c79114f2ec5ba737/source/module_cell/test/parallel_kpoints_test.sh#L1-L2 should use set -e here, since the script is executed as https://github.com/deepmodeling/abacus-develop/blob/332c875f328f3da8975059d4c79114f2ec5ba737/source/module_cell/test/CMakeLists.txt#L96

caic99 commented 4 months ago

The error you proposed is introduced by #4289 (previous run is OK). @dyzheng @mohanchen Any suggestions?

dyzheng commented 4 months ago

@jwsyzy hello, can you look at this Issue?

WHUweiqingzhou commented 4 months ago

@mohanchen could you have a look?