Bug 44640 - REGRESSION (r53420): Parts of aktrailhead.com invisible until page resize
Summary: REGRESSION (r53420): Parts of aktrailhead.com invisible until page resize
Status: RESOLVED CONFIGURATION CHANGED
Alias: None
Product: WebKit
Classification: Unclassified
Component: Layout and Rendering (show other bugs)
Version: 528+ (Nightly build)
Hardware: Mac (Intel) OS X 10.6
: P2 Major
Assignee: Nobody
URL: http://aktrailhead.com
Keywords: InRadar, NeedsReduction, Regression
Depends on: 33245
Blocks:
  Show dependency treegraph
 
Reported: 2010-08-25 15:25 PDT by bugsandstuff
Modified: 2022-02-12 18:28 PST (History)
6 users (show)

See Also:


Attachments

Note You need to log in before you can comment on or make changes to this bug.
Description bugsandstuff 2010-08-25 15:25:29 PDT
For some reason, the newest versions of google chrome sand safari stopped displaying my page correctly when google adsense code is included. Problem started several months ago.

Go take a look at http://aktrailhead.com

It usually loads fine the very first time. Hit refresh (or click any link in the upper green side menu) and wallah, chrome stops loading the page after it renders the adsense code (only displays the menu bar on the left). Why it only does this after reloading or clicking a link, is beyond me.

Safari displays the problem right away, without having to reload. Older versions of chrome (4.x.x) and safari do not exhibit this problem whatsoever. I have replicated this problem on 3 different installations of these browsers on 3 different platforms (Win7, XP, OS 10.6).

My page uses server side includes, if that is any help. My page has passed XHTML strict DOCTYPE tests, and consists of relatively simple code.

See page in FF or IE for comparison.
Comment 1 Alexey Proskuryakov 2010-08-25 16:39:22 PDT
For what it's worth, I cannot reproduce this in Safari/WebKit 5.0.1, nor in Safari with a local build of WebKit, r65737.
Comment 2 bugsandstuff 2010-08-25 20:59:39 PDT
I have replicated this in BOTH chrome and safari, BOTH on WINDOWS and OSX, all up to date!

Chrome: 5.0.375.127
Safari: 5.0.1 (6533.17.18)

That being the case, I am having a mind f*** of a time trying to figure out how anyone else could NOT replicate it.

One common note I have discovered in both operating systems and both browsers is that as soon as I resize the window, the main section of the page (that was missing) appears. ONLY when I re-size. When I click a link or refresh, it goes away until I resize again. BOTH chrome and safari do this (in both OSs).

I haven't been able to get this fixed  for months despite doing nearly every single conceivable thing I can think of of.

Here is every solid FACT I can point out:

1. Affects newer builds of webkit browsers ONLY. Not firefox, not IE, not OLDER chrome or OLDER safari.

3. Only happens with adsense code in the html. Take the adsense code away (and  -nothing else-) and the problem ceases to exist, will always render entire page properly.

2. Happens across multiple platforms and installations - I cannot emphasize that enough! I can even go through the trouble of making screen recorded videos of both browsers in both operating systems, if that's what you want.

Tried removing computer from router, plugged strait in to cable modem too, got ISP-given IP address with no NAT or firewall of any sort. Problem remains.
Comment 3 Alexey Proskuryakov 2010-08-26 10:17:04 PDT
> One common note I have discovered in both operating systems and both browsers is that as soon as I resize the window

Thanks, that was the crucial bit of information needed to reproduce this. Turns out that this only happens if the window is large enough - I was trying to reproduce on a 13" screen, so I didn't see this.

I can now reproduce this problem with Safari 5.0.1, as well as with a local build of r66022.
Comment 4 bugsandstuff 2010-08-26 16:58:36 PDT
> I can now reproduce this problem with Safari 5.0.1, as well as with a local build of r66022.

Alright, that's good to hear! I hope the underlying flaw can be identified and fixed relatively shortly.
Comment 5 Simon Fraser (smfr) 2010-08-26 17:03:50 PDT
s/wallah/voilà
Comment 6 Simon Fraser (smfr) 2010-09-09 13:36:41 PDT
<rdar://problem/8412487>
Comment 7 Andy Estes 2011-01-04 17:55:03 PST
This appears to be a regression caused by http://trac.webkit.org/changeset/53420.
Comment 8 bugsandstuff 2011-01-05 16:08:44 PST
In plain terms, what does this mean exactly?
Comment 9 Alexey Proskuryakov 2011-01-05 16:16:00 PST
This is information for an engineer fixing the bug. Specifically, we now know exactly which change caused the problem.
Comment 10 bugsandstuff 2011-01-05 16:33:31 PST
(In reply to comment #9)
> This is information for an engineer fixing the bug. Specifically, we now know exactly which change caused the problem.

Ah, ok, thanks. Thats good to hear.
Comment 11 Robert Hogan 2013-03-02 12:54:14 PST
I can't reproduce this on Chrome ToT.