Closed Bug 894519 Opened 13 years ago Closed 12 years ago

Intermittent browser_bug882977.js | Identity state should be chromeUI for about:home in a new window - Got unknownIdentity, expected chromeUI

Categories

(Core :: Security, defect)

x86
macOS
defect
Not set
normal

Tracking

()

RESOLVED WORKSFORME
Tracking Status
firefox24 --- unaffected
firefox25 --- affected
firefox26 --- affected

People

(Reporter: RyanVM, Unassigned)

References

Details

(Keywords: intermittent-failure)

https://tbpl.mozilla.org/php/getParsedLog.php?id=25328896&tree=Mozilla-Inbound Rev4 MacOSX Lion 10.7 mozilla-inbound opt test mochitest-browser-chrome on 2013-07-16 07:55:32 PDT for push 3e3d601a9ab2 slave: talos-r4-lion-057 07:59:48 INFO - TEST-START | chrome://mochitests/content/browser/browser/base/content/test/browser_bug882977.js 07:59:48 INFO - TEST-INFO | chrome://mochitests/content/browser/browser/base/content/test/browser_bug882977.js | Console message: OpenGL LayerManager Initialized Succesfully. 07:59:48 INFO - Version: 2.1 NVIDIA-7.12.9 07:59:48 INFO - Vendor: NVIDIA Corporation 07:59:48 INFO - Renderer: NVIDIA GeForce 320M OpenGL Engine 07:59:48 INFO - FBO Texture Target: TEXTURE_2D 07:59:48 WARNING - TEST-UNEXPECTED-FAIL | chrome://mochitests/content/browser/browser/base/content/test/browser_bug882977.js | Identity state should be chromeUI for about:home in a new window - Got unknownIdentity, expected chromeUI 07:59:48 INFO - Stack trace: 07:59:48 INFO - JS frame :: chrome://mochikit/content/browser-test.js :: test_is :: line 536 07:59:48 INFO - JS frame :: chrome://mochitests/content/browser/browser/base/content/test/browser_bug882977.js :: checkIdentityMode :: line 37 07:59:48 INFO - JS frame :: chrome://mochitests/content/browser/browser/base/content/test/browser_bug882977.js :: onLoad :: line 30 07:59:48 INFO - native frame :: <unknown filename> :: <TOP_LEVEL> :: line 0 07:59:48 INFO - INFO TEST-END | chrome://mochitests/content/browser/browser/base/content/test/browser_bug882977.js | finished in 450ms
Blocks: 893424
No longer blocks: 882977
jaws, any idea whether this is a real failure or just the test being fragile?
Flags: needinfo?(jaws)
I don't think the test is being fragile, we were pretty sure to check that the content had finished loading and that the location was correct. I wonder if fixing bug 893937 will fix this orange though.
Flags: needinfo?(jaws)
(In reply to Jared Wein [:jaws] from comment #2) > I don't think the test is being fragile, we were pretty sure to check that > the content had finished loading and that the location was correct. Right, but maybe it's possible for onSecurityChange to be called just a little bit later than that. > I wonder if fixing bug 893937 will fix this orange though. You can still reproduce that with bug 893424's patch?
(In reply to Dão Gottwald [:dao] from comment #3) > (In reply to Jared Wein [:jaws] from comment #2) > > I don't think the test is being fragile, we were pretty sure to check that > > the content had finished loading and that the location was correct. > > Right, but maybe it's possible for onSecurityChange to be called just a > little bit later than that. Yeah, that's a good point. > > I wonder if fixing bug 893937 will fix this orange though. > > You can still reproduce that with bug 893424's patch? Oh, I had assumed that bug 893937 was filed after the patch for 893424 was applied, but I forgot it didn't get merged until this morning and hence isn't out in a Nightly build yet. Please disregard my earlier comment.
Blocks: 882977
Any other ideas, jaws?
Flags: needinfo?(jaws)
We could try two things: 1) In the load event listener, we could try putting some executeSoon's to delay checking the identity state until a bit later. This seems fragile. 2) We could fire a chrome event each time the identity state is changed, and adjust the test to listen for this event. This seems less fragile but also introduces test-only code in to production. Dao, what do you think?
Flags: needinfo?(jaws) → needinfo?(dao)
If one executeSoon helps, I guess that would be fine, but if multiple executeSoons are needed, that seems exceedingly weird and would most likely be fragile. To really understand this, someone would probably take a look at how onSecurityChange works under the hood :/ I agree we should try to avoid adding test-only code.
Flags: needinfo?(dao)
Hard to tell if the intermittent orange is gone, https://tbpl.mozilla.org/?tree=Try&rev=f6c91a2fdc26
(In reply to Jared Wein [:jaws] from comment #53) > Hard to tell if the intermittent orange is gone, > https://tbpl.mozilla.org/?tree=Try&rev=f6c91a2fdc26 Looks encouraging. Both of those failures are known failures unrelated to this.
Closing inactive keywords:intermittent-failure bugs where the TBPLbot has previously commented and the test isn't marked as disabled; filter on orange-cleanup-201401.
Status: NEW → RESOLVED
Closed: 12 years ago
Resolution: --- → WORKSFORME
You need to log in before you can comment on or make changes to this bug.