[Webkit-unassigned] [Bug 11339] Explanation of head vs. body timeout on dead hostnames

[bugzilla-daemon at webkit.org]

http://bugs.webkit.org/show_bug.cgi?id=11339





------- Comment #4 from kurt at formulateaffinity.com  2006-10-19 17:57 PDT -------
I will clarify and attach a file:

Basically, in the head I put this:
<script type="text/javascript" src="http://64.24.56.47/foo.js"></script>
now, that as far as I can tell, and I hope for as long as it needs to be for
this test, is a dead IP, as in not http responds at it at this moment.  It
could be a real IP, or hostname, just not responding due to load etc.

When I load the stall.html file, Activity shows me 0.4KB loaded, which is the
html file itself, Safari shows me nothing, I do not see "this page stalls
badly"

http://64.24.56.47/foo.js shows 0 bytes of ? for a total of:
1 minute and 15 seconds, and then I get 
http://64.24.56.47/foo.js timed out

After that, the rest of the page loads, and I see the body copy.

The problem also happens sometimes, with certain script calls inside the body
as well.  

If you ask Mac users what they think of myspace, they will all say that it is
always "down", whereas windows users do not seem to have the same sentiment. 
While it is true, the .js servers do often go down, that does not prevent a
windows user from loading the page for more than about 15 seconds, from there,
they are ok to move on.  On the Mac, users more or less assume the site is
dead, and will not wait over a minute to see it load.

Yes, the behavior is identical in the nightly, which I just tested in.


-- 
Configure bugmail: http://bugs.webkit.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.