309332 - When one of elements in a page can't be loaded the "Unable to connect"/"The connection was reset" screen overwrites the entire page

Status: RESOLVED DUPLICATE of bug 313519

(Firefox :: General, defect)

Description

[Ryuichi KUBUKI]

Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9a1) Gecko/20050919
Firefox/1.6a1

When one of elements in a page can't be loaded due to connection errors, the
"Unable to connect"/"The connection was reset" screen overwrites the entire page
even though the page is partially loaded. It's mighty annoying to see a
partially loaded page overwritten by this new error message system when one of
banner ads is down.

IMHO it should go back to old modeless error-message dialogue or mini bar
overlayed on a window.

Comment 1

[Ria Klaassen (not reading all bugmail)]

Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9a1) Gecko/20050919
Firefox/1.6a1 ID:2005091922

Can you give an example url for I've never seen this, but maybe I was just lucky.
You could always set browser.xul.error_pages.enabled to false in order to get
the alert back.

Comment 2

[Matt]

www.t-mobile.com does this all the time. After the page appears to be loaded a error comes up that says the connection has been reset.

Comment 3

[Andrew Sterling Hanenkamp]

(In reply to comment #1)
> Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9a1) Gecko/20050919
> Firefox/1.6a1 ID:2005091922
> 
> Can you give an example url for I've never seen this, but maybe I was just lucky.
> You could always set browser.xul.error_pages.enabled to false in order to get
> the alert back.

I saw the same behavior on google.com when my DSL connection flaked out. This bug reminds of the behavior IE exhibits when you place a JavaScript into a <DIV> tag---the page renders, but then disappears when it finishes and tries to run the script. Under no circumstance should a partial rendering cause the "Problem loading page" page to appear. Rather, the little top-bar notice should be used instead (NO POPUPS!).

Comment 5

[Cees T.]

Duplicate of bug 313519?

Comment 6

[Ryuichi KUBUKI]

(In reply to comment #5)
> Duplicate of bug 313519?
> 

OK let's move on...

--> dupe