[Webkit-unassigned] [Bug 27165] Connections-per-host should be tracked closer to the ResourceHandle level, not the Cache

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=27165





--- Comment #37 from Nate Chapin <japhet at chromium.org>  2010-11-05 16:49:20 PST ---
(In reply to comment #36)
> (From update of attachment 72883 [details])
> View in context: https://bugs.webkit.org/attachment.cgi?id=72883&action=review
> 
> >> Did this test pass in Firefox and IE in its original form? If it did, we should probably find a way to keep it working.
> > The data I provided before was for the original form.  I can get it for the new form of the test if you like, but I was mostly pointing out that behavior on this test isn't standard as is.
> 
> So, we used to be compatible with Firefox and IE8, and no longer are. This seems quite unfortunate. Perhaps credentials should be stored immediately, not waiting for a request to actually start? Or maybe requests with explicit credentials should always get highest priority, to make sure that subsequent ones are properly authenticated (or logged out)?
> 
> I don't think this difference necessarily warrants r-, but logging out is such a sensitive area that one gets paranoid.

I just experimented with the suggestion to give High priority to requests with explicit credentials.  It appears that this breaks the test, but in a different way.  The test currently sends an async request, then logs out, and expects the async request to return 200.  With higher priority for explicit credentials, it looks like we end up sending the logout request before the async request.  I'm not sure whether we're worried about that case.

I'd be happy to also try your suggestion of storing credentials immediately, but I'm having trouble finding the code where we store the credentials.  Is that somewhere in WebCore?

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