286130 - 100% CPU usage, page is not rendered.

Status: RESOLVED INVALID

(Firefox :: General, defect)

Description

[pg]

User-Agent:       Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.7.6) Gecko/20050225 Firefox/1.0.1
Build Identifier: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.7.6) Gecko/20050225 Firefox/1.0.1

trying to view the web page (though not w3c conform renders ok in IE) the
browser starts to use 100% CPU, and becomes unresponsive. The only way to deal
with the problem was to kill firefox (loosing all other tabs).
(It's not good, that a page can crash the browser.)


Reproducible: Always

Steps to Reproduce:
view: http://www.mistral.hu/arlista_kozos.php?id_shop=5

Actual Results:  
100% CPU usage, firefox not responding.

Expected Results:  
render the web page or say that it cannot render it.

Comment 1

[Alessandro Minelli]

Testcase seems to be not longer valid...

Comment 2

[Tristor]

The URL provided doesn't appear to be valid.  This needs a valid testcase or
some other way to test for reproduction.  If it causes a crash, can you provide
Talkback IDs?  

Comment 3

[pg]

The site has been changed. (Maybe because of the bug:-)
Now a similar page (with less colums) can be found at:

http://www.mistral.hu/arlista_print.php?id_shop=1

This page renders (ugly) in 39 secons on a 1915 MHz Athlon XP 2600+, so there is
no crash now... (Rendering takes less than a second in IE.)
The page is much smaller than the original one, so maybe if I were more patient
it could render even the original page.

Now I tested using:
Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.7.6) Gecko/20050317
Firefox/1.0.2

Now a harmful web page can DOS attack the browser by sending it a simple page... 

Comment 4

[Tristor]

Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.8b2) Gecko/20050329
Firefox/1.0+

Testing using the new URL provided in comment #3, I did encounter some lag.  I
think it is partially the server I connected to being slow/on the other side of
the planet, however, it might happen to have something to do with the ENOURMOUS
amount of divs in that page.  I looked at the source, that page scares me.  It
does indeed spike the CPU usage up to 100% and sit there until the page is
finished loading/rendering.  If anything, this would be a bug in the rendering
engine when dealing with massive amounts of divs.  I believe this is actually a
duplicate of another bug, so I am going to go search for it now.

Comment 5

[Tristor]

Although I can reproduce the effect reported, it is due to the page being
extremely long and so full of divs that it just takes a long time to render.  At
the recommendation of another QA person, I am going to mark this Invalid.