SANS ISC: False Positive or Not? Difficult to Analyze Javascript - SANS Internet Storm Center

• SANS Site Network
  • Current Site
  • Internet Storm Center
  • Other SANS Sites Help
  • Graduate Degree Programs
  • Security Training
  • Security Certification
  • Security Awareness Training
  • Penetration Testing
  • Industrial Control Systems
  • Cyber Defense Foundations
  • DFIR
  • Software Security
  • Government OnSite Training

SANS ISC InfoSec Forums

Our reader Travis sent us the following message:

We have had 2 users this morning hit a Forbes page: hxxp://www.forbes.com/sites/jimblasingame/2013/05/07/success-or-achievement/

And then after being referred from there to: hxxp://ml314.com/tag.aspx?2772014

They are setting off our FireEye web appliance. It is advising that this is an "Infection Match" which I am not entirely familiar with their systems determinations as it is fairly new to us. I called down the source of the link they went to and can submit that as well if you would like it, but I haven't had a chance to look at it yet just beautified it and saved it.

I went ahead and downloaded the "ml314.com" URL using wget, and what comes back is heavily obfuscated Javascript. I am just quoting some excerpts of it below:

(function(a){var g=window.document;var h=[];var e=[];var f=(g.readyState=="complete"||g.readyState=="loaded"||g.readyState=="interactive");var d=null;var j=function(k){try{k.apply(this,e)}catch(l){if(d!==null){d.call(this,l)}}};var c=functi...36);F=p(F,D,B,G,E[1],12,-389564586);G=p(G,F,D,B,E[2],17,606105819);B=p(B,G,F,D,E[3],22,-1044525330);D=p(D,B,G,F,E[4],7,-176418897);F=p(F,D,B,G,E[5],12,1200080426);G=p(G,F,D,B,E[6],17,-1473231341);B=p(B,G,F,D,E[7],22,-45705983);D=p(D,B,G,F,E[8],7,1770035416);F=p(F,D,B,G,E[9],12,-1958414417);G=p(G,F,D,B,E[10],17,-42063);B=p(B,G,F,D,E[11],22,-1990404162);D=p(D,B,G,F,E[12],7,1804603682);F=p(F,D,B,G,E[13],12,-40341101);G=p(G,F,D, ... function f(o){o.preventDefault();o.stopPropagation()}function i(o){if(g){return g}if(o.matches){g=o.matches}if(o.webkitMatchesSelector){g=o.webkitMatchesSelector}if(o.mozMatchesSelector){g=o.mozMatchesSelector}if(o.msMatchesSelector){g=o.msMatchesSelector}if(o.oMat ... try{s=new ActiveXObject("ShockwaveFlash.ShockwaveFlash");p=s.GetVariable("$version").substring(4);p=p.split(",");p=p[0]+"."+p[1]}catch(r){}if(s){q="Flash"}return{name:q,version:

In short: Very obfuscated (not just "minimized"), and a lot of keywords that point to detecting plugin versions. Something that you would certainly find in your average exploit kit. But overall, it didn't quite "add up". Not having a ton of time, I ran it through a couple Javascript de-obfuscators without much luck. The domain "ml314.com" also looked a bit "odd", but lets see when it was registered:

$ whois ml314.comâ??

   Domain Name: ML314.COM
   Name Server: NS.RACKSPACE.COM
   Name Server: NS2.RACKSPACE.COM
   Updated Date: 22-apr-2013
   Creation Date: 22-apr-2013
   Expiration Date: 22-apr-2018

â??Admin Organization: Madison Logic
Admin Street: 257 Park Ave South
Admin Street: 5th Floor

The domain name isn't new, and hosted in what I would call a "decent" neighborhood on the Internet. The owner information doesn't look outright fake, and indeed gives us a bit more information to solve the puzzle. Turns out that "Madison Logic" is in the web advertisement / click through business, so what you are seeing is likely their proprietary Javascript to track users better. 

In the end, I call this a "false positive", but then again, feel free to correct me. This is just one example how sometimes things are not simple "black/white" when it comes to odd Javascript.

---
Johannes B. Ullrich, Ph.D.
STI|Twitter|LinkedIn

I will be teaching next: Intrusion Detection In-Depth - SANS Madrid March 2019

I have been seeing similar requests, although the response from ml314.com is much longer, more elaborate and again checks for plugins like Silverlight and Windows Media Player.

I would agree that this is almost certainly advertising of some sort, but with my experience of malware infested ad servers, I am choosing to block it.

We've had a ticket open with FireEye support on this to determine whether or not it's a false positive since it began firing heavily on Wednesday. We haven't received an answer yet, however.

Fireeye support just notified us that it's a false positive detection that was removed 8/27/2017 at 12:00 PM PST in SC 348.120.