609718 - Assertion in browser_webconsole_chrome.js

Status: RESOLVED WORKSFORME

(DevTools :: Console, defect, P2)

Description

[David Dahl :ddahl]

Attachment: Assertion bt

I se an assertion in this test:

browser_webconsole_chrome.js

###!!! ASSERTION: Oops!  You're asking for a weak reference to an object that doesn't support that.: 'factoryPtr', file nsWeakReference.cpp, line 109


All checks pass, however.

Comment 1

[David Dahl :ddahl]

This may only be with the lazy console patches from bug 568629 and the integration patch in bug 587734

Comment 2

[David Dahl :ddahl]

Comment on attachment 488309 [details]
Assertion bt

Perhaps we should CC someone who might know this code well? Of course this code is Copyright (C) 1999 code:)

Comment 3

[:Gavin Sharp [email: [e-mail]]]

Are you sure the assertion is caused by hudservice code and that test specifically? That stack looks like nsIObserverService::removeObserver() being called from a pagehide handler (possibly caused by a tab close?), and I don't see any code doing that in HUDService.jsm or in the test.

Comment 4

[David Dahl :ddahl]

(In reply to comment #3)
> Are you sure the assertion is caused by hudservice code and that test
> specifically? That stack looks like nsIObserverService::removeObserver() being
> called from a pagehide handler (possibly caused by a tab close?), and I don't
> see any code doing that in HUDService.jsm or in the test.

I was unsure, and saw this several times during test runs. It may not be hudservice code. Thanks for looking into this.

Comment 5

[David Dahl :ddahl]

Also, I only noticed this after the test checks were finished and mochitest was about to load the next test.

Comment 6

[Kevin Dangoor]

mass change: filter on PRIORITYSETTING

Comment 7

[David Dahl :ddahl]

This may have to do with the observers shutting down on shutdown

Comment 8

[Mihai Sucan [:msucan]]

Is this still valid? I would be inclined to say no...

Comment 9

[Rob Campbell [:rc] (:robcee)]

Unlikely.