1047846 - Delayed iFrame Request Fails

Status: UNCONFIRMED

(Core :: DOM: Core & HTML, defect, P5)

Description

[jud]

User Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_9_2) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/36.0.1985.125 Safari/537.36

Steps to reproduce:

1. On page load, we load the DFP GPT tag asynchronously.

2. After the page loads, approx. 5-10 seconds, we create the first look 728x90 ad slot and then immediately display it in the appropriate div on the page. All ad slots are created with enableSingleRequest turned off.

3. After the preroll finishes (30-45 seconds), we create the first look 300x250 ad slot and attempt to display it. The ad fails to display in Firefox as DFP/GPT does not make a request for an ad. 


Actual results:


It seems there are issues accessing IFRAME properties. Hacking around the dfp JS it seems uf might be trying to see if a child iFrame exists but cannot tell for sure.

Broken in FF v. 31+ and suspected to have started in FF 30. Works in FF =>v29 and obviously works in all other browsers. Effectively stops any delayed display advertising using Google's DFP GPT tags and likely affecting other executions... Creates an error in GPT uf=null as seen in the Google Publisher Tools.

Broken: http://screencast.com/t/Xxt57KwE



Expected results:

We've used this execution for a number of years now and no changes were made on our side or on DFP/GPT from what we can tell.

Working v29: http://screencast.com/t/qSF5BXRX0f

DFP/GPT script should not error and does not except in FF v31+ tested v32 and same issue exists.

Comment 1

[jud]

In the screenshots the 300x250 unit on the left where you see "Ad Unit Did Not Fetch", and "Ad Unit Did Not Render" is an Ad unit exhibiting symptoms of the issue. Down below in the same screenshot you can find the cause with the (!) in red.

Comment 2

[Boris Zbarsky [:bzbarsky]]

Can you please link to a page that shows the problem?  It's ... hard to debug a screencast, since that doesn't show what your scripts are doing, what your HTML looks like, or anything else relevant other than the existence of the behavior.

Comment 3

[jud]

http://player.radioloyalty.com/station/3131.html

Comment 4

[Boris Zbarsky [:bzbarsky]]

Sorry for the delay here...

I see nothing like what your screenshots are showing, in either Firefox 31 or Firefox 30 (though I'm not looking at this in Google Publisher Tools, just in the browser).

In fact, the behavior I see is identical in Firefox 30 and Firefox 31.

I would really appreciate some steps that I can use to actually reproduce the problem, or a clear description of how to tell whether I'm reproducing the problem if just loading the page is supposed to be enough to reproduce it.

Comment 5

[Firefox Bug Husbandry Bot]

https://bugzilla.mozilla.org/show_bug.cgi?id=1472046

Move all DOM bugs that haven’t been updated in more than 3 years and has no one currently assigned to P5.

If you have questions, please contact :mdaly.