RESOLVED FIXED 50529
Citibank website, loads then goes completely blank 
https://bugs.webkit.org/show_bug.cgi?id=50529
Summary Citibank website, loads then goes completely blank
Ryosuke Niwa
Reported 2010-12-04 15:07:59 PST
Reproduction steps 1. Go to citicards.com 2. Enter Account Login Info 3. Wait for it to go completely white.. What is the expected result? For the account details to not dissappear What happens instead? The entire page goes blank
Attachments
Add attachment proposed patch, testcase, etc.
Ryosuke Niwa
Comment 1 2010-12-04 15:08:54 PST
http://crbug.com/60038
Alexey Proskuryakov
Comment 2 2010-12-05 01:04:00 PST
Is there any output in console? I don't have a Citybank account, and thus can't reproduce. Is this a regression? If it's a recent one, it could be related to bug 50526.
Ryosuke Niwa
Comment 3 2010-12-06 14:18:44 PST
(In reply to comment #2) > Is there any output in console? I don't have a Citybank account, and thus can't reproduce. > > Is this a regression? If it's a recent one, it could be related to bug 50526. It could be but I can reproduce it on both WebKit TOT and Safari.
James Robinson
Comment 4 2010-12-06 15:45:42 PST
*** This bug has been marked as a duplicate of bug 50526 ***
Ryosuke Niwa
Comment 5 2010-12-06 16:04:40 PST
(In reply to comment #4) > > *** This bug has been marked as a duplicate of bug 50526 *** This is not a duplicate of 50526 because the bug 50529 reproduces on r72487 but the bug 50526 doesn't In fact, the bug is regression between r72954 and r72991.
Ryosuke Niwa
Comment 6 2010-12-06 23:26:18 PST
On inspector, body has display: none, which is set by a script. This might be their bug but it's weird that it only reproduces on WebKit.
Tony Gentilcore
Comment 7 2010-12-07 08:33:32 PST
(In reply to comment #6) > On inspector, body has display: none, which is set by a script. This might be their bug but it's weird that it only reproduces on WebKit. If it has display:none, this isn't the late doc.write() blowing away the document bug. However, due to the raciness, it is still likely to be related to script execution order. I still haven't had any luck reproducing this. Have you tried changing the user agent? It is possible one of the scripts is doing UA string detection and taking different paths based on what it thinks the browser supports.
Ryosuke Niwa
Comment 8 2010-12-22 10:06:22 PST
(In reply to comment #7) > If it has display:none, this isn't the late doc.write() blowing away the document bug. However, due to the raciness, it is still likely to be related to script execution order. I still haven't had any luck reproducing this. Have you tried changing the user agent? It is possible one of the scripts is doing UA string detection and taking different paths based on what it thinks the browser supports. I tried Internet Explorer and Firefox but neither worked.
Tony Gentilcore
Comment 9 2010-12-29 12:10:57 PST
Although I haven't reduced, symptoms like this could likely be explained by: https://bugs.webkit.org/show_bug.cgi?id=8852
Ryosuke Niwa
Comment 10 2011-01-10 00:40:18 PST
The bug still reproduces on r75294.
Tony Gentilcore
Comment 11 2011-01-10 09:23:14 PST
(In reply to comment #10) > The bug still reproduces on r75294. rniwa: If you File > Save Page As in chrome, then open from local disk, does it still repro? If so, start ripping things out until it doesn't repro to see if you can pinpoint the script responsible. I'd love to get to the bottom of this but simply can't repro with my account (I've tried on 4 different machines).
Alexey Proskuryakov
Comment 12 2011-01-20 17:01:22 PST
I agree that this really needs a reduction.
Ryosuke Niwa
Comment 13 2011-02-20 04:03:02 PST
Citicards "fixed" their website and the bug doesn't reproduce anymore.
Note You need to log in before you can comment on or make changes to this bug.