Difference between revisions of "Testing for a successful setup"

From Kraken Wiki
Jump to: navigation, search
(Created page with " After going through the set up steps you can check for a successful installation with this approach. ==== Checking solver environment variabl...")
 
(Running tests)
Line 41: Line 41:
 
All of the tests should be completed successfully, i.e. you should see <tt>Test completed with exit code 0</tt> as the verdict for each test and a final result of <tt>All tests passed</tt>.
 
All of the tests should be completed successfully, i.e. you should see <tt>Test completed with exit code 0</tt> as the verdict for each test and a final result of <tt>All tests passed</tt>.
  
If you have any failed tests you can check the output of that tests in  
+
If you have any completely failed tests (<tt>Test terminated with exit code 1</tt>) you can check the output of that tests in  
  
 
  tests/workspace/<test_number>/test_output.txt
 
  tests/workspace/<test_number>/test_output.txt

Revision as of 10:15, 30 January 2024

After going through the set up steps you can check for a successful installation with this approach.

Checking solver environment variables

Put the path of each solver into environment variables in the Kraken environment.

Test that the environment variable paths work in the Kraken environment by simply running (in the environment):

$ANTS_EXE_PATH

This should provide some output from Ants:

Usage: ants [OPTION] INPUT-FILE

You can similarly test

$SUPERFINIX_EXE_PATH
$KHARON_EXE_PATH

and

$SERPENT_EXE_PATH

if you have Serpent installed.

Running tests

Each of the solvers and the Python packages have their own folder containing some simple regression tests utilizing the testing module of KrakenTools:

  • KrakenTools/tests/
  • Cerberus/tests/
  • Cetus/tests/
  • Ants/tests/
  • Kharon/tests/
  • SuperFINIX/tests/

You can navigate to these test folders and run the tests using

python runTests.py

All of the tests should be completed successfully, i.e. you should see Test completed with exit code 0 as the verdict for each test and a final result of All tests passed.

If you have any completely failed tests (Test terminated with exit code 1) you can check the output of that tests in

tests/workspace/<test_number>/test_output.txt

to try to diagnose the problem.

If you have tests timing out and the test output indicates that the test did not simply hang you can adjust the timeout in the runTests.py file.