Closed Bug 755127 Opened 14 years ago Closed 14 years ago

mozmill python tests broken on HEAD

Categories

(Testing Graveyard :: Mozmill, defect)

x86
Linux
defect
Not set
normal

Tracking

(Not tracked)

RESOLVED FIXED

People

(Reporter: k0scist, Unassigned)

References

Details

Attachments

(1 file)

Now that the python tests actually take the binary, running them is giving me lots of errors: FAILED (failures=4, errors=1) INFO | Some tests were unsuccessful. INFO | INFO | +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ INFO | INFO | Python Failures: INFO | INFO | (<testapi.TestMozmillAPI testMethod=test_api>, 'Traceback (most recent call last):\n File "/home/jhammel/mozmill/src/mozmill/mutt/mutt/tests/python/testapi.py", line 24, in test_api\n self.assertTrue(len(results.passes) == passes)\nAssertionError: False is not true\n') INFO | INFO | (<testpersisted.TestMozmillPersisted testMethod=test_persisted>, 'Traceback (most recent call last):\n File "/home/jhammel/mozmill/src/mozmill/mutt/mutt/tests/python/testpersisted.py", line 38, in test_persisted\n self.assertTrue(len(results.passes) == 1)\nAssertionError: False is not true\n') INFO | INFO | (<testpersisted.TestMozmillPersisted testMethod=test_persisted_shutdown>, 'Traceback (most recent call last):\n File "/home/jhammel/mozmill/src/mozmill/mutt/mutt/tests/python/testpersisted.py", line 48, in test_persisted_shutdown\n self.assertTrue(len(results.passes) == 1)\nAssertionError: False is not true\n') INFO | INFO | (<testmultiplerun.TestMozmillAPI testMethod=test_runtwice>, 'Traceback (most recent call last):\n File "/home/jhammel/mozmill/src/mozmill/mutt/mutt/tests/python/testmultiplerun.py", line 24, in test_runtwice\n results = m.run(dict(path=path))\n File "/home/jhammel/mozmill/src/mozmill/mozmill/mozmill/__init__.py", line 352, in run\n self.run_tests(*tests)\n File "/home/jhammel/mozmill/src/mozmill/mozmill/mozmill/__init__.py", line 319, in run_tests\n frame = self.start_runner()\n File "/home/jhammel/mozmill/src/mozmill/mozmill/mozmill/__init__.py", line 251, in start_runner\n self.runner.start()\n File "/home/jhammel/mozmill/src/mozbase/mozrunner/mozrunner/runner.py", line 122, in start\n self.profile.reset()\n File "/home/jhammel/mozmill/src/mozbase/mozprofile/mozprofile/profile.py", line 98, in reset\n proxy = self._proxy)\n File "/home/jhammel/mozmill/src/mozbase/mozprofile/mozprofile/profile.py", line 78, in __init__\n self.addon_manager.install_addons(addons, addon_manifests)\n File "/home/jhammel/mozmill/src/mozbase/mozprofile/mozprofile/addons.py", line 49, in install_addons\n self.install_from_path(addon)\n File "/home/jhammel/mozmill/src/mozbase/mozprofile/mozprofile/addons.py", line 177, in install_from_path\n assert os.path.isdir(path), "Addon \'%s\' cannot be installed" % path\nAssertionError: Addon \'/tmp/tmpqPD6Yg.mozrunner/extensions/mozmill@mozilla.com\' cannot be installed\n') INFO | INFO | (<test_bug690154.TestBug690154 testMethod=test_JSON_structure>, 'Traceback (most recent call last):\n File "/home/jhammel/mozmill/src/mozmill/mutt/mutt/tests/python/test_bug690154.py", line 34, in test_JSON_structure\n self.assertTrue(fails[\'name\'] == \'<TOP_LEVEL>\')\nKeyError: \'name\'\n') INFO | INFO | =========================================================================== INFO | INFO | No Javascript Failures INFO | FWIW, I am getting massive failures on the JS tests too (not shown here), but one thing at at time This could be https://bugzilla.mozilla.org/show_bug.cgi?id=696468 but I'm not sure yet. Will dig in as I am able
Just to be curious, why are failures on Python shown via the INFO logger level? Those should be ERROR.
The output is from the summary at the end, as the entire output is very long. I didn't design mutt, though I assume the rationale is that the final summary should be printed at the INFO level. As to whether this is "correct" or not, I don't really have a strong opinion and consider the issue orthogonal to the failures themselves
Attached file sample manifest
So I've been playing with a reduced manifest as given in attachment 624221 [details] to try to debug these one by one. The trick is....the failures are related. If I run usemozmill.py or testapi.py, they pass. Albeit, I get INFO | Timeout: bridge.execFunction("39e0c3b2-9ede-11e1-a11c-0024d7065b60", bridge.registry["{6d856535-6cdc-4ced-ad6d-145ca11cacc7}"]["cleanQuit"], []) for these and any other test I run o_O. This is probably the actual problem. However, if I run testapi.py following usemozmill.py, as given in the manifest, I get (output truncated): INFO | Timeout: bridge.execFunction("13518506-9ede-11e1-a11c-0024d7065b60", bridge.registry["{8d5da637-544d-416a-b8ef-e7e7a84136f4}"]["cleanQuit"], []) INFO | INFO | Passed: 1 INFO | Failed: 0 INFO | Skipped: 0 .INFO | Timeout: bridge.set("38532af8-9ede-11e1-a11c-0024d7065b60", Components.utils.import('resource://mozmill/driver/mozmill.js')) INFO | INFO | Timeout: bridge.execFunction("39e0c3b2-9ede-11e1-a11c-0024d7065b60", bridge.registry["{6d856535-6cdc-4ced-ad6d-145ca11cacc7}"]["cleanQuit"], []) INFO | F ====================================================================== FAIL: test_api (testapi.TestMozmillAPI) ---------------------------------------------------------------------- Traceback (most recent call last): File "/home/jhammel/mozmill/src/mozmill/mutt/mutt/tests/python/testapi.py", line 24, in test_api self.assertTrue(len(results.passes) == passes) AssertionError: False is not true ---------------------------------------------------------------------- Ran 2 tests in 137.264s FAILED (failures=1) INFO | Running JS Tests INFO | INFO | Some tests were unsuccessful. INFO | INFO | +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ INFO | INFO | Python Failures: INFO | INFO | (<testapi.TestMozmillAPI testMethod=test_api>, 'Traceback (most recent call last):\n File "/home/jhammel/mozmill/src/mozmill/mutt/mutt/tests/python/testapi.py", line 24, in test_api\n self.assertTrue(len(results.passes) == passes)\nAssertionError: False is not true\n') INFO | Note the timeout in loading resource://mozmill/driver/mozmill.js . Seems common sense that if we can't load this file that horrible things will happen. I'm guessing that this is due to the timeout on cleanQuit in the previous test. To date, I have no idea what causes it. To repeat, running testapi.py in isolation does pass, albeit with the same timeout on cleanQuit
:ctalbert, do you get similar? Or anyone else running linux?
See Also: → 696468
On a computer where sample.ini does run correctly (see comments in bug 696468) the only failure I see is: ====================================================================== FAIL: test_runtwice (testmultiplerun.TestMozmillAPI) ---------------------------------------------------------------------- Traceback (most recent call last): File "/home/jhammel/mozmill/src/mozmill/mutt/mutt/tests/python/testmultiplerun.py", line 24, in test_runtwice results = m.run(dict(path=path)) File "/home/jhammel/mozmill/src/mozmill/mozmill/mozmill/__init__.py", line 352, in run self.run_tests(*tests) File "/home/jhammel/mozmill/src/mozmill/mozmill/mozmill/__init__.py", line 319, in run_tests frame = self.start_runner() File "/home/jhammel/mozmill/src/mozmill/mozmill/mozmill/__init__.py", line 251, in start_runner self.runner.start() File "/home/jhammel/mozmill/src/mozbase/mozrunner/mozrunner/runner.py", line 162, in start self.profile.reset() File "/home/jhammel/mozmill/src/mozbase/mozprofile/mozprofile/profile.py", line 98, in reset proxy = self._proxy) File "/home/jhammel/mozmill/src/mozbase/mozprofile/mozprofile/profile.py", line 78, in __init__ self.addon_manager.install_addons(addons, addon_manifests) File "/home/jhammel/mozmill/src/mozbase/mozprofile/mozprofile/addons.py", line 49, in install_addons self.install_from_path(addon) File "/home/jhammel/mozmill/src/mozbase/mozprofile/mozprofile/addons.py", line 177, in install_from_path assert os.path.isdir(path), "Addon '%s' cannot be installed" % path AssertionError: Addon '/tmp/tmpzGIyYb.mozrunner/extensions/mozmill@mozilla.com' cannot be installed ---- I recall having seen a bug to this effect. Will look into it
Since Bug 753190, things seem to be working. I see an intermittent failure in expectstacktest.py but cannot reliably reproduce. Will close this and subsequent issues can be ticketed subsequently
this also doesn't account for bug 696468 to be tracked there
Status: NEW → RESOLVED
Closed: 14 years ago
Resolution: --- → FIXED
Product: Testing → Testing Graveyard
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Creator:
Created:
Updated:
Size: