m86 Web Exploits Report
m86 Web Exploits Report
m86 Web Exploits Report
EXECUTIVE SUMMARY
In the last few years M86 Security Labs has seen a dramatic increase in attack or exploit kits. These easy-to-use kits are the
backbone of exploits in the “wild”. M86 Security Labs research reviews how exploit kits are developed, distributed and monetized
globally. The turnover of exploits is quick. The success rate is high. And, all of this for very minimal cost for the exploit kit users
and operators. The details in this report will provide a fundamental understanding of how exploits operate and give the reader a
true sense of the business behind the crime.
In the Internet security industry, the terms “exploit kit” or “attack toolkit” are commonly known and understood by security
researchers. However, to the average Internet user, these exploit kits are unfamiliar. So, what exactly are these tools? Why are
they written? Who uses them and what makes them so popular -- especially, in the wrong hands?
Figure 1 illustrates the login page for one of the newest toolkits available today. It glorifies cybercrime as a serious business,
showing images of money, drugs and a gun to convey the typical rewards you can expect when you use the “crimepack”
exploit kit.
The main motivation driving the cybercrime industry is the possibility of monetary gain. Cybercriminals find it easier, faster and
more cost effective to make money by buying exploits rather than taking the time to create exploits themselves. The demand for
these types of tools drives opportunities. Savvy, knowledgeable individuals with skills in developing Web applications and basic
knowledge in hacking have filled a niche by creating exploit kits.
An exploit kit is a Web application that is developed using web technologies such as PHP and database products such as
MySQL. They allow a kit user to take advantage of the most known exploits in popular applications, such as Microsoft’s Internet
Explorer, Adobe Acrobat, Reader and Flash Player, as well as many others. The kit is installed on a web server somewhere
connected to a database for logging and reporting. The kit interfaces are web based, as you can see in the early example of Web
Attacker in Figure two. Cheap, free and highly anonymous web hosting is easily available today and many Cybercriminals take
advantage of these types of services to host their exploit kits.
m86security.com
Exploit kits began appearing in early 2006. The first known, popular exploit kit was Web Attacker, which exploited
seven previously known vulnerabilities targeting Internet Explorer and Mozilla Firefox browsers. Before exploit kits were
available, Cybercriminals had to craft and develop their own exploits, which involved research and testing to discover new
vulnerabilities, a highly technical task.
The availability of Web Attacker opened the floodgates to newer exploit kits, such as MPack and GPack. As these kits grew
in popularity, new exploit kits began to appear more frequently. In the last six months, M86 Security Labs has observed at
least a dozen new kits being used in the wild.
Exploit kits are routinely being advertised in underground hacker forums. Often, the interfaces to these kits are in Russian,
with English being used sparingly indicating perhaps, their target market.
Figures 5 and 6 display examples of advertisements for exploit kits. Figure 5 shows an advertisement for the Fragus exploit
kit that supports a multilingual interface, while Figure 6 highlights the Crimepack exploit kit which promises the “highest
(exploit) rates for the lowest price.”
Sometimes the professional nature of these exploit kits can be seen in their own dedicated Web sites (Figure 7).
Purchasers of exploit kits can expect to receive free services from the authors, such as continuing support for their kits, bug
fixes, minor version changes and other small features in just the same way as legitimate software companies.
The heart and soul of the exploit kits are the exploits themselves. The kits typically employ the most well known, published
exploit code. Most often, we see unchanged proof of concept code from security related sites or forums, ranging from fairly
old to the latest zero-day exploits. Beyond browser-based exploits, M86 Security Labs has noticed an increase in popularity
of Adobe Flash, Java classes, and PDF-based exploits.
In the case of malicious Javascript code, it is almost always obfuscated, greatly reducing the effectiveness of classic
signature-based security products.
When a kit is successfully deployed, its payload is often a Trojan horse that is downloaded to the victim’s machine. Often,
we have seen that malicious code isn’t provided twice for the same user (IP address), complicating a security researcher’s
forensic work. Regardless of whether the initial attack is successful or not, the same attack usually will not be duplicated for a
second attempt because the results will remain the same. If the attack was successful there is no reason to supply malicious
code a second time, if it was not successful there is still no reason to attack. This evasive behavior prevents a security
researcher from analyzing the code to craft a signature/rule to prevent it.
The user can also view statistics using the Web-based interface. The kind of statistics available to the user varies from the
number of successful exploits, the victim’s geographical location, the operating system, and browser type and version being used
by their victim, among other details.
With only a few hundred dollars, anyone can become the proud owner of one of the latest versions of an exploit kit. This wasn’t
difficult until now, right? It’s interesting to note that at this point, there are very few difficulties faced by the novice user of an
exploit kit, even one with minimum technical skills.
In order to configure the exploit kit, the user must install it onto a Web server. It’s very easy to find free hosting, or for minimal
cost, paid hosting can provide the standard set of services needed to install an exploit kit, such as PHP and MySQL.
Translation:
Installation:
- Upload the files to the server
- Upload your file to the same folder and rename it to 1.exe
- Create DB
- Open config.php and add db information, full path to pfg.php and full path to load.php,
Also password/login for statistics access (password should be encrypted using MD5 twice!)
- Now browse to http://your_site/spl/_install.php
If installation succeeded, you will see “Installation finished Please delete install.php”
The most technical information a user needs to know are basic Unix commands for copying a file (“cp”), modifying file permissions
(“chmod”), and removing a file (“rm”). Knowing these three simple commands can result in a successful installation of an exploit
kit. In some cases, even this knowledge is not necessary, as some toolkits can be fully installed through a Web-based interface
as can be seen in Figure 14.
One of the most important measures of every exploit kit is the percentage of successful exploits, also known as the Exploitation
Rate. This rate depends on several parameters, such as the type of traffic or quality of the exploit code, and the ability to
customize exploits for every victim (based on browsers and/or other client application versions).
The most important criteria is the selection of available exploits that are used by the exploit kit. Most kits provide a different set
of exploits for different browsers-- from the antiquated MDAC exploit for Internet Explorer 6, to the Holy Trinity of infamous PDF
exploits (printf, collectEmailInfo and getIcon) which affect the large user base of Adobe Acrobat/Reader users. Of course, the best
option for successful exploitation is zero-day exploits. Most often, the exploit kit creators continually update the set of exploits
included in their product to maintain a high exploitation rate.
Translation:
Version 1.3.2 (16.12.2009)
• pdf Doc.media.newPlayer added (currently 0-day <= 9.2 (Adobe Reader version)
• Pdf Crypting changed
• Java calendar added
Version 1.3.1 (16.11.2009)
• Blocking SE bots (search engines) functionality added. Reduce the risk of domain
disclosure
-Robots.txt added
• Pdf was changed a little
• Java D&E renewed (increases exploitation rate, works for opera)
Version 1.3 (25.10.2009)
…
Pdf provided only if vulnerable Adobe acrobat version installed
…
Version 1.2 (27.07.2009)
…
Snapshot removed
Spreadsheet added
…
Version 1.1
DirectX DirectShow exploit added
Tag Font exploit added, exploiting Firefox 3.5
…
Version 1.0 (release)
Exploits:
MDAC
MS009-02
Snapshot
Telnet – for opera
PDF collab.getIcon
PDF Util.Printf
PDF collab.collectEmailInfo
Some exploit kits do allow the user to choose the set of exploits to be implemented, but generally they are preconfigured.
After the successful deployment of an exploit kit, the only remaining task is how to direct the largest possible number of
victims to the kit’s exploit page. This is a fundamental problem faced by the user of the kit. Like any business, the exploit
buyer seeks to maximize its exposure and subsequently the revenue. So, how does the attacker solve the problem? There
are a few options: the first is to utilize sending spam messages with an appropriate link (often known as a blended threat
email), another popular method is to create bogus Web sites and promote them through various search engine optimization
(SEO) techniques. However, the most effective technique is injecting iFrames within a legitimate Web site that direct back to
the exploit page. The iFrame injection is possible by attacking legitimate Web servers (for example, using SQL injection) or if
available, using stolen FTP credentials. Exploit kit operators might carry out these activities themselves, or pay someone else
for these services.
In order to increase the exploitation rate, a user needs people to visit their exploit page. To achieve this , there are individuals
who specialize in selling Web traffic for any purpose.
Translation:
Selling Traffic iframe .ru
Traffic iframe .ru
Price for 1K / 3 WMZ (web money)
Working with protection!
Up to 50K in a day!
ICQ …
Cybercriminals will select traffic which is most suitable for their planned criminal activity. For example, if the user plans to drop
banking Trojans onto victims’ machines to steal money from their bank accounts, the Cybercriminal will prefer traffic from
wealthy Western countries where there is a higher chance to find people using online banks.
STATISTICS
From this point, the exploit kit’s malicious page will be provided to the victim without their knowledge. The law of large
numbers, which holds that even a small percentage of a large number is still a large number, allows novice criminals create a
small-scale botnet in a fairly short time.
After acquiring an exploit kit, the chief goal of the cybercriminal is to make money, and there are numerous ways this may
be achieved. But first, it’s important to understand that the operators of exploit kits are merely one part of an extensive
underground economy where the participants are often specialized, offering tailored products and services to other players
through shady forums and personal contacts.
Cybercriminals are interested in using exploit kits to install malware for personal gain and potential profitability. The kit
operators may install their own creations or third-party malware. In most cases, the installed malware is usually a version of a
bot client which enables the bot herder to control the infected host for the following purposes:
• Stealing critical information from the victim, e.g. keyloggers or other malware attacks where the stolen data is later sold or
used.
• Using the victims’ computing resources for sending spam, where the bot herder earns money for messages sent, or by
signing up to a spam affiliate program like the common “Canadian Pharmacy” program.
• Installing other malware like fake anti-virus scareware, where revenues can be earned from successful “registrations”, or
Pay-Per-Install (PPI) programs.
One popular example of a cybercriminal’s method of making money is Pay-Per-Install (PPI) programs, where the criminals
are paid for installing third-party malware. In this case, the exploit kit operator finds a suitable PPI program and becomes an
affiliate. Affiliate members obtain malware from the PPI program’s Web site and get paid for each successful install of the
malware on a victim’s computer. PPI programs are prolific and varied; some PPI programs create their own malware, while
others are merely distributing third-party malware. The malware itself can vary, ranging from information stealers like Zeus,
spambots like Rustock, password stealers, or generic downloaders. The exploit kit operator might also bundle several pieces
of malware from different PPI programs in the payload.
Payments of successful installs are made regularly to affiliates, often on a daily or monthly basis. The rates per install vary
depending on what country the target computer is located. The United States is a favored country, where installs command a
higher price. Below is a recent pay schedule for one PPI site:
Other types of payment programs also exist. Notable are those programs dealing with fake anti-virus ‘scareware’ products,
which trick users into paying ‘registration’ fees to enable the ‘protection’ software. These programs have a revenue sharing
model, where affiliates are paid a share of the revenue generated from the registration fees originating from their installs. Some
sites advertise up to a 60% share of these revenues.
In order to give life to some of the concepts above, let’s consider a real life example, which we observed in February 2010 through
a spam campaign touting ‘photoshock’ pictures. The operator of the exploit kit most likely paid another party to perform the spam
campaign for them, and in this case the spam originated from the Pushdo/Cutwail botnet.
In this case the landing page contained a hidden iFrame that allowed exploits to be served up from another server hosting the
FS Pack kit.
Figure 23: Hidden iFrame Pulling in Content from a Remote Server Hosting an Exploit Kit
The admin pages from the exploit kit clearly show the zinnko.com referrer domains used in the spam campaign, as well as
another campaign using facebook.com in the URLs.
The admin page in Figure 25 shows 5,032 successful installs for the day. Assuming a PPI model where the affiliate is earning
a modest $100.00 USD per 1,000 installs, this would result in revenue of about $500.00 USD for the day.
Figure 25: FS Pack Admin Console Showing Number of Successful Installs at 5,032
In this paper we have looked at the history of exploit kits or attack toolkits, seen examples of what they look like, how they work
and discussed what they can be used for. Also of importance was the point on just how little technical knowledge an aspiring
Cybercriminal needs to become active.
The second part of the paper looked at the money trail. We discovered how much these kits are sold for, therefore how much
their creators are making. We then went down a level and looked at the kit operators and the different options they have to make
money. We presented detail on one method, Pay-Per-Install or PPI programs and how the operators were either paid for each
successful install or received a share of any revenues through affiliate programs.
The aim of this paper was to explain the exploit kits, how they work and how easy they are to use. It hopefully gives some insight
into why we are seeing such a massive increase in the number of attacks targeting exploits and to what we are facing in today’s
Internet threat landscape.
• WebAttacker • Fiesta
• MPack • Eleonore
• GPack • MyLoader
• AdPack • SEO Toolkit
• IcePack • JustExploit Elite Loader
• Neosploit • Clean Pack
• MyPolySploit • Shamans Dream
• XCore • Siberia
• UniquePack • Fragus
• LuckySploit • Max Toolkit
• Yes Toolkit • CrimePack
• SPack • FSPack
• Liberty • and others
M86 Security is the global expert in real-time threat protection and the industry’s leading Secure Web Gateway provider. The
company’s appliance, software, and Software as a Service (SaaS) solutions for Web and email security protect more than 24,000
customers and over 17 million users worldwide. M86 products use patented real-time code analysis and behavior-based malware
detection technologies as well as threat intelligence from M86 Security Labs to protect networks against new and advance
threats, secure confidential information, and ensure regulatory compliance. The company is based in Orange, California with
international headquarters in London and development centers in California, Israel, and New Zealand.
© Copyright 2010 M86 Security. All rights reserved. M86 Security is a registered trademark of M86 Security. All other product and company names mentioned herein are trademarks or registered trademarks of their respective companies.