Last modified: 2013-08-30 11:13:02 UTC

Wikimedia Bugzilla is closed!

Wikimedia migrated from Bugzilla to Phabricator. Bug reports are handled in Wikimedia Phabricator.
This static website is read-only and for historical purposes. It is not possible to log in and except for displaying bug reports and their history, links might be broken. See T48234, the corresponding Phabricator task for complete and up-to-date bug report information.
Bug 46234 - Selenium tests that drive Internet Explorer are not stable enough
Selenium tests that drive Internet Explorer are not stable enough
Status: VERIFIED WONTFIX
Product: Wikimedia
Classification: Unclassified
Quality Assurance (Other open bugs)
unspecified
All All
: Unprioritized normal (vote)
: ---
Assigned To: Nobody - You can work on this!
:
Depends on:
Blocks:
  Show dependency treegraph
 
Reported: 2013-03-16 22:36 UTC by Željko Filipin
Modified: 2013-08-30 11:13 UTC (History)
5 users (show)

See Also:
Web browser: ---
Mobile Platform: ---
Assignee Huggle Beta Tester: ---


Attachments

Description Željko Filipin 2013-03-16 22:36:54 UTC
Internet Explorer (6-9) tests fail randomly. Investigate the failures and make the tests more robust.
Comment 1 Ori Livneh 2013-04-01 01:38:35 UTC
Here's one run that seems bogus:
https://wmf.ci.cloudbees.com/job/browsertests-windows-internet_explorer_9/267/testReport/(root)/Math/Display_simple_math/

Sauce job:
https://saucelabs.com/jobs/302b2cf196374819b39470b2b3e55918#

Raw selenium log:
https://saucelabs.com/jobs/302b2cf196374819b39470b2b3e55918/selenium-server.log

The failure is:
org.openqa.selenium.NoSuchElementException: Unable to find element with id == wpTextbox1 (WARNING: The server did not provide any stacktrace information)

..but if you look at the video, #wpTextbox is clearly there.

Interestingly, the first NoSuchElementException failure is thrown *before* another warning, "Invalid length: Content-Length=22200 written=8192". Although this warning is emitted for load.php (i.e., for one of the ResourceLoader requests and and not for the main request), it makes me suspect that Selenium is not waiting for DOMContentLoaded, and that there is consequently a race condition that causes the selector to fire some milliseconds before the element is present.

If this is indeed the problem, it should be fixable by setting a wait timeout.
See http://www.deanhume.com/Home/BlogPost/selenium-webdriver---wait-for-an-element-to-load/64 & http://docs.seleniumhq.org/docs/04_webdriver_advanced.jsp.
Comment 2 Ori Livneh 2013-04-01 01:40:09 UTC
..also, it might be a good idea to disable wmf-selenium-bot until this bug is fixed, since the bogus failures are not winning it any friends.
Comment 3 Željko Filipin 2013-04-05 10:11:12 UTC
Since just 1-2 tests are failing on Internet Explorer, disable them until they are fixed. Failing tests now just create noise, and the vast majority of the tests are passing.
Comment 4 Željko Filipin 2013-04-05 10:12:11 UTC
Oli, I have spend a few hours trying to fix failing IE tests, but I had no luck so far. I will disable failing tests for now (for IE only), until I get some time to make them more robust.
Comment 5 Željko Filipin 2013-04-05 10:12:35 UTC
"Oli" should be "Ori", of course. :)
Comment 6 Željko Filipin 2013-04-08 17:29:31 UTC
Failing tests disabled in https://gerrit.wikimedia.org/r/#/c/58099/
Comment 7 Željko Filipin 2013-08-30 11:10:31 UTC
Wontfix-ing, this is not a useful bug, it is too general. I will create more targeted bugs as needed.

Note You need to log in before you can comment on or make changes to this bug.


Navigation
Links