Closed Bug 796911 Opened 12 years ago Closed 12 years ago

Intermittent test_clear_browser_data.html | localstorage data have been deleted - got bar, expected null

Categories

(Core :: General, defect)

x86
Linux
defect
Not set
normal

Tracking

()

RESOLVED FIXED
mozilla19
Tracking Status
firefox18 --- fixed

People

(Reporter: emorley, Assigned: mounir)

References

Details

(Keywords: intermittent-failure)

Attachments

(4 files)

(Test newly added by bug 795134)

Rev3 Fedora 12 mozilla-inbound opt test mochitest-other on 2012-10-01 06:10:29 PDT for push 144db3479fe3

slave: talos-r3-fed-061

https://tbpl.mozilla.org/php/getParsedLog.php?id=15702195&tree=Mozilla-Inbound

{
9318 INFO TEST-START | chrome://mochitests/content/chrome/dom/tests/mochitest/localstorage/test_clear_browser_data.html
9319 INFO TEST-PASS | chrome://mochitests/content/chrome/dom/tests/mochitest/localstorage/test_clear_browser_data.html | no data - null should equal null
9320 INFO TEST-PASS | chrome://mochitests/content/chrome/dom/tests/mochitest/localstorage/test_clear_browser_data.html | data written - bar should equal bar
9321 INFO TEST-PASS | chrome://mochitests/content/chrome/dom/tests/mochitest/localstorage/test_clear_browser_data.html | no data - null should equal null
9322 INFO TEST-PASS | chrome://mochitests/content/chrome/dom/tests/mochitest/localstorage/test_clear_browser_data.html | data written - bar should equal bar
9323 INFO TEST-PASS | chrome://mochitests/content/chrome/dom/tests/mochitest/localstorage/test_clear_browser_data.html | no data - null should equal null
9324 INFO TEST-PASS | chrome://mochitests/content/chrome/dom/tests/mochitest/localstorage/test_clear_browser_data.html | data written - bar should equal bar
9325 INFO TEST-PASS | chrome://mochitests/content/chrome/dom/tests/mochitest/localstorage/test_clear_browser_data.html | no data - null should equal null
9326 INFO TEST-PASS | chrome://mochitests/content/chrome/dom/tests/mochitest/localstorage/test_clear_browser_data.html | data written - bar should equal bar
9327 INFO TEST-PASS | chrome://mochitests/content/chrome/dom/tests/mochitest/localstorage/test_clear_browser_data.html | no data - null should equal null
9328 INFO TEST-PASS | chrome://mochitests/content/chrome/dom/tests/mochitest/localstorage/test_clear_browser_data.html | data written - bar should equal bar
9329 INFO TEST-PASS | chrome://mochitests/content/chrome/dom/tests/mochitest/localstorage/test_clear_browser_data.html | no data - null should equal null
9330 INFO TEST-PASS | chrome://mochitests/content/chrome/dom/tests/mochitest/localstorage/test_clear_browser_data.html | data written - bar should equal bar
9331 ERROR TEST-UNEXPECTED-FAIL | chrome://mochitests/content/chrome/dom/tests/mochitest/localstorage/test_clear_browser_data.html | localstorage data have been deleted - got bar, expected null
9332 INFO TEST-PASS | chrome://mochitests/content/chrome/dom/tests/mochitest/localstorage/test_clear_browser_data.html | sessionstorage data have not been deleted - bar should equal bar
9333 INFO TEST-PASS | chrome://mochitests/content/chrome/dom/tests/mochitest/localstorage/test_clear_browser_data.html | data are still there - bar should equal bar
9334 INFO TEST-PASS | chrome://mochitests/content/chrome/dom/tests/mochitest/localstorage/test_clear_browser_data.html | data are still there - bar should equal bar
9335 INFO TEST-PASS | chrome://mochitests/content/chrome/dom/tests/mochitest/localstorage/test_clear_browser_data.html | data are still there - bar should equal bar
9336 INFO TEST-PASS | chrome://mochitests/content/chrome/dom/tests/mochitest/localstorage/test_clear_browser_data.html | data are still there - bar should equal bar
9337 INFO TEST-END | chrome://mochitests/content/chrome/dom/tests/mochitest/localstorage/test_clear_browser_data.html | finished in 734ms
}
Attached patch Try #1Splinter Review
Hopefully it's just the checks running too early (before the frame calls clearBrowserData().
Unfortunately, the app frame is quite isolated and being able to inform the parent something is done wouldn't be trivial.
Attachment #667410 - Flags: review?(bmo)
Comment on attachment 667410 [details] [diff] [review]
Try #1

Thank you for the quick patch :-)

Unfortunately this isn't my area of expertise so I wouldn't feel comfortable reviewing it - sorry!
Attachment #667410 - Flags: review?(bmo)
(In reply to Ed Morley [:edmorley UTC+1] from comment #3)
> Unfortunately this isn't my area of expertise so I wouldn't feel comfortable
> reviewing it - sorry!

You can't file random orange and not be able to fix or review patches ;) I will be glad to help you dig into our test framework.
Attachment #667410 - Flags: review?(jwatt)
Attachment #667410 - Flags: review?(jwatt) → review+
(In reply to Mounir Lamouri (:mounir) from comment #4)
> You can't file random orange and not be able to fix or review patches ;) I
> will be glad to help you dig into our test framework.

It's on my list of many things to learn more about; unfortunately for now at least there is so much of a backlog of unowned sheriffing tasks that I just haven't had time yet. Once tooling and such is improved, sheriffing itself will be less of a burden so I will hopefully have more time for things like this :-)
Attachment #667410 - Flags: checkin+
https://hg.mozilla.org/integration/mozilla-inbound/rev/1e5584c8e7d0
Status: NEW → ASSIGNED
Flags: in-testsuite+
Target Milestone: --- → mozilla18
https://hg.mozilla.org/mozilla-central/rev/1e5584c8e7d0
Status: ASSIGNED → RESOLVED
Closed: 12 years ago
Resolution: --- → FIXED
Status: RESOLVED → REOPENED
Resolution: FIXED → ---
Attached patch DebugSplinter Review
Just to make sure the issues is what I think it is.
Attached patch Try #2Splinter Review
I believe that should work but I really wonder why I wasn't able to access the inner frame's variable from the outer frame but I'm able to access the DOM. Boris, do you happen to have an idea why?
Attachment #669623 - Flags: review?(bzbarsky)
Which inner iframe variable could you not access?  How were you accessing it?
(In reply to Boris Zbarsky (:bz) from comment #21)
> Which inner iframe variable could you not access?  How were you accessing it?

I was doing:
 window.foo = "bar";
in the iframe, at load time.

and:
 alert(frames[1].foo);
in the outer iframe (inner iframe's event handler). That was showing "undefined". Same if the outer iframe was setting a variable in the inner iframe.

That's why I'm using the DOM to say that the call has been done...
That should absolutely have worked.  Care to post a non-working version of the patch so I can take a look?
That patch isn't working (timing out). Am I doing something wrong Boris?
> That patch isn't working (timing out). Am I doing something wrong Boris?

Ah.  So I applied this locally and tried to run it, at which point I realized it's a chrome test, not a mochitest.

So the test page is running as chrome.  But the subframe is not chrome, so you get an Xray, and Xrays hide expandos.  Using "frames[1].wrappedJSObject.cleared !== true" as the test makes it work.  But I guess using the DOM is fine too.
Comment on attachment 669623 [details] [diff] [review]
Try #2

r=me
Attachment #669623 - Flags: review?(bzbarsky) → review+
https://hg.mozilla.org/integration/mozilla-inbound/rev/49b130f753d3
Status: REOPENED → ASSIGNED
Target Milestone: mozilla18 → mozilla19
Attachment #669623 - Flags: checkin+
Comment on attachment 669129 [details] [diff] [review]
Debug

I reverted that patch while pushing try #2.
Attachment #669129 - Flags: checkin+ → checkin-
Comment on attachment 667410 [details] [diff] [review]
Try #1

This is no longer applied (replaced by Try #2).
Attachment #667410 - Flags: checkin+ → checkin-
https://hg.mozilla.org/mozilla-central/rev/49b130f753d3
Status: ASSIGNED → RESOLVED
Closed: 12 years ago12 years ago
Resolution: --- → FIXED
Whiteboard: [orange]
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Created:
Updated:
Size: