175941 - Browser hangs and consumes 100% CPU time on URL

Status: RESOLVED WORKSFORME

(SeaMonkey :: General, defect)

Description

[Mark Grosberg]

User-Agent:       Mozilla/5.0 (Macintosh; U; PPC Mac OS X; en-US; rv:1.2b) Gecko/20021016
Build Identifier: Mozilla/5.0 (Macintosh; U; PPC Mac OS X; en-US; rv:1.2b) Gecko/20021016

I am not sure if this is a dupe or not. The HTML on this page is horrid. The
site seems to use JavaScript for an animated background (I don't think this was
a flash plug-in issue).

When navigating to the front page the browser responds to paint events but not
to anything else (it selects the busy mouse pointer). It consumes 100% of the
CPU time. 

I did some searches for dup's but didn't find any relating to this URL and the
HTML may be worth investigation. 

This page worked fine in MSIE. 

If this turns out to be just a general DOM/JS perf issue the browser should at
least not display the "I'm busy doing something" mouse cursor.

This was on a 500MHz G4 under OS X.2. No other apps except Terminal were active
at the time. 
 

Reproducible: Always

Steps to Reproduce:
1. Go to URL
2. Watch Mozilla lock. During this state it will repaint the display but not
respond to menus.
3. Only resort is to kill browser from shell.

Actual Results:  
Browser hung.

Expected Results:  
Rendered the page as IE had or given an error. Spinning the busy cursor is not
intuitive.

Comment 1

[Bryan Blackburn]

There's a javascript in it called Starfield from
http://www.btinternet.com/~kurt.grigg/javascript which appears to be doing it. 
Going to this site, I didn't see anything by that exact name, but "Mouse Stars"
seems to mess up Mozilla as well (2002102003).

Comment 2

[Greg K.]

If someone could, please capture some Sampler data from Mozilla during this hang
and attach it to this bug.

Comment 3

[Bryan Blackburn]

Attachment: Data from running "sample 4528 10 50"

Requested output from using sample while Mozilla was going crazy.

Comment 4

[Greg K.]

Seems to WorkForMe using FizzillaMach/2003021203. Mark, can you still reproduce
this problem using a current nightly build?

Comment 5

[Mark Grosberg]

Sorry for the delay, been a bit busy...

No, I have not seen this bug on any later releases. However, it was not easy to
reproduce on the effected build (which I am using now). I tried some of the 1.3
betas and couldn't reproduce it. Of course, it's hard for me to duplicate the
workload my machine was under when I filed this bug.

Let's WFM it and I'll file a new one if I see it in a later release.

Comment 6

[Greg K.]

Resolving WFM per comment 5.