921959 - nsIWebNavigation::reload(LOAD_FLAGS_ALLOW_MIXED_CONTENT) fails in e10s

Status: RESOLVED FIXED

(Core :: DOM: Navigation, defect, P2)

Description

[Mark Hammond [:markh] [:mhammond]]

browser_bug822367.js and browser_bug902156.js both are mixed-content tests and call reload() with LOAD_FLAGS_ALLOW_MIXED_CONTENT.  These tests fail and the test output includes:

0:08.33 TEST-INFO | chrome://mochitests/content/browser/browser/base/content/test/general/browser_bug902156.js | Console message: [JavaScript Error: "NS_ERROR_ILLEGAL_VALUE: Component returned failure code: 0x80070057 (NS_ERROR_ILLEGAL_VALUE) [nsIWebNavigation.reload]" {file: "chrome://global/content/browser-child.js" line: 163}]

Instrumenting browser-child.js shows that it only seems to fail when the flags are LOAD_FLAGS_ALLOW_MIXED_CONTENT.  Haven't dug any deeper than this!

Comment 1

[Tanvi Vyas[:tanvi]]

LOAD_FLAGS_ALLOW_MIXED_CONTENT is used when a user clicks on the shield doorhanger and decides to disable mixed content blocking protection on a page.  When the user does this, we reload the page with this flag.  When we detect this flag on reload, we set something called the MixedContentChannel to the current channel:
https://mxr.mozilla.org/mozilla-central/source/docshell/base/nsDocShell.cpp#9634

When deciding whether to allow or block mixed content, we compare the MixedContentChannel against the current channel to see if the user disabled protection for the page:
https://mxr.mozilla.org/mozilla-central/source/docshell/base/nsDocShell.cpp#5505

Unfortunately, I won't be able to look into these e10s failures anytime soon.

Comment 2

[Jim Mathies [:jimm]]

Attachment: enabled tests

Appears we fixed this at some point, these tests pass for me with --e10s.

Comment 3

[Jim Mathies [:jimm]]

try push:
https://tbpl.mozilla.org/?tree=Try&rev=73b9344e47e2

Comment 4

[Jim Mathies [:jimm]]

(In reply to Jim Mathies [:jimm] from comment #3)
> try push:
> https://tbpl.mozilla.org/?tree=Try&re..

push w/properly formatted --e10s flags - 

https://tbpl.mozilla.org/?tree=Try&rev=a69f03edd2d8

Comment 5

[Jim Mathies [:jimm]]

(In reply to Jim Mathies [:jimm] from comment #4)
> (In reply to Jim Mathies [:jimm] from comment #3)
> > try push:
> > https://tbpl.mozilla.org/?tree=Try&re..
> 
> push w/properly formatted --e10s flags - 
> 
> https://tbpl.mozilla.org/?tree=Try&rev=a69f03edd2d8

Hrm, unfortunately none of the bc runs got far enough to actually run these tests. They should show up in bc1.

Comment 6

[Jim Mathies [:jimm]]

Attachment: patch for checkin

Comment 7

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/integration/mozilla-inbound/rev/eaffaa176816

Comment 8

[Carsten Book [:Tomcat]]

https://hg.mozilla.org/mozilla-central/rev/eaffaa176816