lp://staging/~javier.collado/utah/returncodes
- Get this branch:
- bzr branch lp://staging/~javier.collado/utah/returncodes
Branch merges
- UTAH Dev: Pending requested
-
Diff: 391 lines (+75/-35)11 files modifieddebian/changelog (+3/-0)
examples/run_install_test.py (+4/-3)
examples/run_test_bamboo_feeder.py (+4/-3)
examples/run_test_cobbler.py (+4/-3)
examples/run_test_vm.py (+4/-3)
examples/run_utah_tests.py (+5/-4)
utah/client/common.py (+15/-8)
utah/client/result.py (+1/-4)
utah/client/runner.py (+4/-4)
utah/client/tests/test_runner.py (+2/-2)
utah/run.py (+29/-1)
Related bugs
Bug #1025633: Standardize return codes | Wishlist | Fix Released |
|
Bug #1092638: utah client exits with status 0 when testsuite setup fails | Medium | Fix Released |
|
Bug #1133227: Use different exit status for provisionning errors and test errors | Medium | Fix Released |
|
Related blueprints
Branch information
Recent revisions
- 838. By Javier Collado
-
Refactored client return codes to use a class instead of a dictionary
Suggested by Andy
- 837. By Javier Collado
-
Refactored server return codes to use a class instead of a dictionary
Suggested by Andy
- 836. By Javier Collado
-
Clarified return code values in server and client
(LP: #1025633, LP: #1133227) - 835. By Javier Collado
-
Added server return codes dictionary
The server has its own return codes and an offset that is added to the client
return code when an error is returned to make sure it's possible to know when
an error came from the client or the server just looking at the return code. - 833. By Javier Collado
-
Removed error on test case command failure
Any non success returncode (!= 0) in the test case command is considered a
failure. Before this change, if the test case command returned 1 it was
considered a failure and an error otherwise. Since no assumption can be made on
the returncode from the test case command, this change considers that
everything is a failure.Note that an non successful returncode in the setup command is always
considered and error. Hence, the sum of errors and fetch_errors is the number
of test cases that weren't really executed because something failed before the
command was executed while the number of failures is the number of test cases
commands that failed for some reason. - 832. By Javier Collado
-
Removed processes return code constants
These constants are not being used anywhere else, so it makes sense to remove
them and avoid confusing them with the return codes used by the client. - 831. By Javier Collado
-
Hardcoded user unknown return code
Instead of using a constant that isn't using anywhere else, this change
replaces it with the constant 1 which happens to be the same code that sudo
would have returned when trying to execute.I think this is better to avoid confussing return codes from commands executed
by the client and return codes from the client itself.
Branch metadata
- Branch format:
- Branch format 7
- Repository format:
- Bazaar repository format 2a (needs bzr 1.16 or later)
- Stacked on:
- lp://staging/utah