Closed Bug 1156427 Opened 10 years ago Closed 8 years ago

Intermittent test_navigation.py TestNavigate.test_set_location_through_execute_script | AssertionError: 'http://127.0.0.1:49295/test.html' != u'about:blank'

Categories

(Testing :: Marionette Client and Harness, defect)

defect
Not set
normal

Tracking

(firefox49 wontfix, firefox50 fixed, firefox51 fixed)

RESOLVED FIXED
mozilla51
Tracking Status
firefox49 --- wontfix
firefox50 --- fixed
firefox51 --- fixed

People

(Reporter: RyanVM, Assigned: whimboo)

References

Details

(Keywords: intermittent-failure, pi-marionette-intermittent)

Attachments

(1 file)

20:28:21 INFO - TEST-START | test_navigation.py TestNavigate.test_set_location_through_execute_script 20:28:21 ERROR - TEST-UNEXPECTED-FAIL | test_navigation.py TestNavigate.test_set_location_through_execute_script | AssertionError: 'http://127.0.0.1:49295/test.html' != u'about:blank' 20:28:21 INFO - Traceback (most recent call last): 20:28:21 INFO - File "C:\slave\test\build\venv\lib\site-packages\marionette\marionette_test.py", line 296, in run 20:28:21 INFO - testMethod() 20:28:21 INFO - File "C:\slave\test\build\tests\marionette\tests\testing\marionette\client\marionette\tests\unit\test_navigation.py", line 18, in test_set_location_through_execute_script 20:28:21 INFO - self.assertEqual(self.test_doc, self.location_href) 20:28:21 INFO - TEST-INFO took 281ms
OS: Windows 8.1 → All
Hardware: x86_64 → All
The problem here is that setting window.location.href doesn't wait for the page to be loaded. So instead of using assertEqual() we have to use Wait().until() in this case.
Assignee: nobody → hskupin
Status: NEW → ASSIGNED
Comment on attachment 8789380 [details] Bug 1156427 - Fix test_set_location_through_execute_script by waiting for the page to be loaded. https://reviewboard.mozilla.org/r/77638/#review77102
Attachment #8789380 - Flags: review?(dburns) → review+
Pushed by hskupin@mozilla.com: https://hg.mozilla.org/integration/autoland/rev/10322a5fc3e7 Fix test_set_location_through_execute_script by waiting for the page to be loaded. r=automatedtester
Blocks: 1297394
This is strange. This failure was not visible to me when I tested the patch locally. Maybe something else changed since then which is it causing to fail now. I will have a look and sorry for the perma failures.
Ok, so I figured out the problem. I was creating this patch on my seconds machine (which is a MBP). On that machine I had an old Nightly build I was running the test against. After updating Firefox to latest Nightly I can also see this failure locally. So my assumption from last comment was correct. I did a quick regression range check and as it looks like we regressed between 08/04 and 08/05. Here the appropriate changelog: https://hg.mozilla.org/mozilla-central/pushloghtml?fromchange=1576e7bc1bec7232e9e4ba78cce62526b1a6380b&tochange=d320ef56876f52db9bc0eb79554c7332d4793769 I assume that the patch from Andreas on bug 1280300 could be the regressor here.
Blocks: 1302707
Comment on attachment 8789380 [details] Bug 1156427 - Fix test_set_location_through_execute_script by waiting for the page to be loaded. https://reviewboard.mozilla.org/r/77638/#review77300
Attachment #8789380 - Flags: review?(ato) → review+
Pushed by hskupin@mozilla.com: https://hg.mozilla.org/integration/autoland/rev/79bebd89f7ec Fix test_set_location_through_execute_script by waiting for the page to be loaded. r=ato,automatedtester
Status: ASSIGNED → RESOLVED
Closed: 8 years ago
Resolution: --- → FIXED
Target Milestone: --- → mozilla51
Whiteboard: [checkin-needed-aurora]
Flags: in-testsuite+
Whiteboard: [checkin-needed-aurora]
Backed out from Aurora for e10s test_navigation.py timeouts. https://treeherder.mozilla.org/logviewer.html#?job_id=3561629&repo=mozilla-aurora TEST-UNEXPECTED-ERROR | test_navigation.py TestNavigate.test_set_location_through_execute_script | TimeoutException: TimeoutException: Timed out after 5.5 seconds https://hg.mozilla.org/releases/mozilla-aurora/rev/132a7234541c
Flags: needinfo?(hskupin)
Interesting. Given that this failure has a very low occurrence on mozilla-aurora I don't think it makes sense to invest more time for investigation. Lets leave it as it is.
Flags: needinfo?(hskupin)
Going to re-land this on 50 since the failures this got backed out for are hitting elsewhere anyway, so we'll at least have a consistent failure mode across all branches this way.
Product: Testing → Remote Protocol
Moving bug to Testing::Marionette Client and Harness component per bug 1815831.
Component: Marionette → Marionette Client and Harness
Product: Remote Protocol → Testing
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Created:
Updated:
Size: