Posts categorized “Uncategorized”

Targeted Not Targeted

In the past, I used to encounter people who consistently expressed skepticism about truly targeted activity “why would a nation-state target us?” Following the onslaught of publicly available reports of APT activity over the years, I more commonly encounter those that interpret indiscriminate malicious activity encountered by their organization as targeted “we were targeted by group/malware!” Both takes make it difficult to assess and prioritize the threats organizations face in a meaningful way. Sometimes, I think that the terminology we use gets in the way of precisely conveying the targeting preferences, or lack thereof, when it comes to threat activity.

Targeted vs. Personalization

Mass malware distribution operations may implement automated ways to personalize emails in an attempt to create legitimacy and lure the recipient into opening a malicious file or link. Sometimes, this is fairly simple (adding the domain name from the recipient’s email address to the subject line, looking up the domain name to find an organization name and appending that to the subject line etc.) and in other cases leveraging stolen email threads or email accounts using data obtained from previous compromises. As a result, some mass distribution activity may appear targeted, but is in fact personalized using automation.

This doesn’t mean that a malware delivery campaign – whether personalized or not – cannot leverage an opportunity to evolve into much more targeted activity. This is where understanding the nature of the initial attack vector and likely “follow-on” activity (h/t Cian, are you even on the Twitter?) can inform your defense — both posture and response. Emotet, for example, may drop TrickBot which may be followed by Powershell Empire or Cobalt strike followed by network wide deployment of ransomware, such as Ryuk.

Understanding the nature of the attack vector (not everything is “spearphishing” T1193) and what behavior to expect in a post-compromise situation allows for a more accurate assessment of what tactical remediation steps need to be taken next and what strategic defensive measures should be put in place.

Targeted vs. Impacted

Most vendors produce quarterly or yearly reports that rely on telemetry data to depict the top threats based on the highest volume of detections, often segmented by industry vertical or geographic location. But higher volumes concentrated within specific industry verticals or geographic regions do not necessarily indicate targeting – these volumes are influenced by a variety of factors. @tiskimber and I gave a presentation on this topic at the FireEye Cyber Defence Summit in 2018.

In short, due to the mechanisms by which organizations collect or compile telemetry data as well as the type and location of the sensors, reported statistics can be distorted. Moreover, the highest volume threats will not necessarily be the most significant for an organization because truly targeted threats often involve customized, low-volume attacks.

Therefore, I prefer to use the term impacted to describe the volume/breadth of telemetry data observed within an industry vertical or a geographic region. And I use the term targeted when analysis indicates that an observed threat was used by a specific threat actor group known to conduct targeted attacks and/or was used in a campaign specific to a geographic or industry vertical.

Targeted Not Targeted

The ability to assess whether an alert is related to truly targeted activity or a personalized mass distribution campaign informs your prioritization and response. Knowing what is specifically targeting – as opposed to indiscriminately impacting — your organization, your industry vertical or your region allows you to more accurately evaluate and prioritize threats.

“Commodity Malware” is not the Opposite of Targeted Malware

I really don’t like the term “commodity malware”. It’s not that commodity is necessarily an inaccurate description of a particular piece of malware. And it’s not that knowing whether malware can be purchased or is publicly available is not useful, it’s that just because malware may be characterized as commodity it does not indicate whether or not its being used in a targeted manner.

Over the years numerous publicly available RATs (e.g. PoisonIvy, Gh0st, DarkComet, XtremeRAT etc.) have been used by by emerging cyber-espionage actors (e.g. Syria) as well as long standing ones (e.g. China). This does not preclude these actors from using exclusive, custom malware as well, its just another option for them.

For emerging actors, it may be a cost effective way to jump start cyber-espionage capability and for well-established ones it may be a way to hinder attribution efforts. In either case, there’s a wide variety of options ranging from HackForums to Hacking Team.

Whether malware is available for sale or not doesn’t seem to be the reason why certain malware gets labeled as commodity or not. I don’t recall the malware sold by Hacking Team or FinFisher being routinely described as “commodity” yet exclusive cybercrime operations (e.g. Dridex) are sometimes described as commodity malware. While there may be affiliate relationships, not anyone can just buy such malware.

“Unlike most malware distributors, the Bugat/Dridex enterprise maintains tight control over the Bugat/Dridex malware code and does not appear to sell or distribute it to anyone outside the organization.”

Sometimes, I think what people really mean when using the term is whether the malware is typically distributed in an indiscriminate or targeted manner, and not whether that malware can be purchased or not.

From a defenders perspective, knowing that certain malware is exclusive to a particular threat group that conducts targeted attacks allows you to prioritize and respond to such incidents quickly. If the malware is used by a variety of actors, some of whom conduct targeted activity and some that engage in indiscriminate activity then it requires an additional assessment to determine what type of actor is most likely involved in any particular case.

But even that is murky.

The most recent incarnation of this blurring of indiscriminate and targeted activity has been exemplified by the use of ransomware deployed after an initial, indiscriminate compromise. In these examples, Trickbot and Dridex compromises are followed by 1) interactive activity leveraging Red team tools (such as Powershell Empire, which are typically not described as ‘commodity”) and 2) the deployment of ransomware (e.g. Ryuk).

Some malware that is freely available, or that can be purchased, is rarely referred to as commodity and some malware that is exclusively used, or tightly controlled within a limited set of actors, is often called commodity. Furthermore, targeted activity can involve freely available malware and indiscriminately distributed malware can quickly turn into targeted activity.

So, I’ll ask again.

10 Years Since Ghostnet

On March 28, 2009 the Citizen Lab released “Tracking GhostNet“. So much has changed since then, both for me personally as well as the research community, the industry and the threat landscape itself.

It has been a long time since I updated this blog, in fact, the last entry was at the end of 2010. The “writing” page has largely been kept up to date with the major papers I’ve contributed to and I continued publicly blogging from 2011 – 2013 at Trend Micro and and at FireEye since then. I’m not really totally sure why I stopped blogging here, but after seeing Ron Deibert and some of my old Citizen Lab colleagues the other day — and we realized that it has literally been 10 years since GhostNet – I’m feeling a bit inspired.

Ron Deibert covered it in Black Code, but I remember crunching through pcaps with Greg Walton, the ones he collected from the Dalai Lama’s Office and other locations. We spotted all the Enfal stuff quickly and eventually we found the beacons for the malware (we probably should have named it :)) which lead to “GhostNet”.

After a little bit of the infamous Google searching…

… all you had to do was visit “/Serverlist.php” on any of the C2 servers (which were obtained from analyzing additional malware samples) and you could see panel.

Soon, Google (2010) would reveal that it had been compromised in what became known as Operation Aurora and “APT” and “Cyber Kill Chain” soon become mainstream. There was an increasing focus on a lot of cyberespionage groups, and on Comment Crew in particular with the notable releases of McAfee’s Shady RAT report (2011) and eventually Mandiant’s blockbuster APT1 report (2013).

Producing public technical papers detailing cyber-espionage activity became a fairly regular occurrence. I documented a lot of the research that influenced me during that time frame in these posts:

Looking Back

Looking back, I think there’s some things we got right with GhostNet, but some that definitely could have been done better.

My biggest regret is that we should have been crystal clear from the outset that there was no “hack back” or anything like that. I spent the next few years trying to clarify what had happened.

I think we did a good job of referencing prior work, in particular the work of Maarten Van Horenbeeck (which had a big impact on me, thanks for the heads-up Oxblood!) and Mikko Hyppönen and the folks at F-Secure.

There were two analyses of the GhostNet malware that I included in the footnotes of the report, but had to be redacted because the command and control servers were still up (and cached in Google) allowing anyone to grab all the victim data:

I regret not reaching out to them, as well as others, and working in a more collaborative way with the broader targeted threats research community. I think this would have really helped in other areas that I think we could have done better:

  • My malware analysis skills were pretty rudimentary at that point (in fact I would still say that I’m not that good and I’m learning from the amazing people I work with all the time).
  • I should have better understood and explained that there were multiple, separate attackers on the same box. Not doing so caused a lot of confusion between what was GhostNet and what were clusters of Enfal activity.
  • We could have handled victim notification better. I think being connected to the research community would have really helped. And we did learn from this, it was great to work with Shadowserver and Steven Adair on the next report.

One of the areas that I think we focused on, but that did always get the attention it deserved, was the importance of field work. This was our version of incident response engagements. Gaining an understanding — even if rudimentary — of the context of what happened in a particular incident, what the attackers did post-compromise and why certain data was stolen, which specific victims were targeted/compromised is extremely important. Greg Walton’s role here cannot be understated.

Finally, I think we handled attribution in a responsible way. We assessed the data that we had and explored alternative scenarios. We discussed freelancing, third-party actors, tacit state-encouragement and the possibility of false flags. We expressed an element of confidence in our suggestion that the “evidence tilts the strongest” toward Chinese state involvement.

Looking back I think the report withstands the test of time.

Looking Forward

Over the years I think there has been a certain level of APT fatigue. The research community broadened and we all began looking at the same things and rushing to publish first (myself included). There seemed to be a backlash in reaction to these reports ranging from “it’s all a bunch of marketing” to “it’s always China”.

Then there was the use of the APT label to deflect responsibility when compromises occurred. Simultaneously, the distinction between the all powerful APT and the lowly “commodity” malware emerged. I’ve never liked this distinction. Gh0st, PoisonIvy and many other publicly available malware families and utilities have been used by both cyberespionage and cybercrime actors of varied skillfulness. The same is true in the modern era with the usage of Red Team frameworks (Metasploit, Cobalt Strike, Powershell Empire) as well as a wide variety of RATs. Dismissing whole swaths of activity, is not probably the best security posture.

I’ve only been sporadically researching cyberespionage since about 2016, and I have largely focused on cybercrime. But I have been following the work of a lot of solid researchers, both new and old school, that are continuing to produce amazing research year after year.

To me, and correct me if I’m wrong, it seems like it’s even harder these days. These are not entirely new developments, but dealing with deliberate attempts by threat actors to mislead on attribution and sorting through the “badtribution” out there present challenges. In addition, I think we’ll see more throw away operations where the things we’re used to clustering on, like command and control servers, won’t be re-used thus reducing the hard overlaps available. And the use of large scale distribution that obscures the targeted nature of post-compromise activity — especially when there’s overlap between traditional cybercrime activity with what seems to be more targeted activity — can further complicate the ability to track and assess the motivations and capabilities of these actors.

Well, I’ll leave it at that, and hopefully I won’t wait years to post again :)

2010 and Beyond

The year of 2010 has been an interesting for malware researchers. From the attacks on Google through to the ShadowNet there have been many interesting cases that targeted high profile targets. However, traditional threats such as Zeus, Spyeye and fake antivirus software continue to be what most Internet users face on a daily basis. Moreover, while attacks that are motivated by politics and espionage are increasing, money continues to be the primary driving force in the malware ecosystem. Here’s my thoughts on some of the trends I’ve focused on this year that we can expect to continue into 2011.

Political motivated DoS attacks
Denial of service attacks continue to be used in order to deny access to web sites at critical times. While the attacks by Anonymous in support of Wikileaks (see Arbor’s analysis here and here) have received much media attention, the website of Wikileaks was attacked just prior to the release of leaked diplomatic cables. However, as the Berkman Center has documented, (distributed) denial of service attacks against non-governmental and independent media continue with an alarming frequency. These attacks are aimed at disabling access to key information resources at specific points in time.

My colleagues Deibert and Rohozinski argue that “[d]isabling or attacking critical information assets at key moments in time—during elections or public demonstrations, for example—may be the most effective tool for influencing political outcomes in cyberspace.” In order to achieve this level of “on demand” disruption, those behind the attacks often outsource these types of attacks to botnets for hire thus blurring the boundaries between cybercrime and politically motivated attacks. We can expect to see a continuation of politically motivated DoS attacks in 2011.

The year of 2010 began with the attacks on Google, dubbed Operation Aurora, which dramatically increased awareness of targeted malware attacks and signified that it is acceptable, and even prudent, that companies disclose such attacks. In fact, some companies began including warnings about such attacks in their SEC filings. However, it is not just companies that are the targets of such attacks, human rights organizations and government systems are compromised as well. In April 2010, the Information Warfare Monitor and the Shadowserver Foundation released a report “Shadows in the Cloud: An investigation into cyber espionage 2.0” in which we document a targeted malware network that extracted secret, confidential and restricted documents from the Indian government and military. (While this report was a follow-up to our previous report on cyber-espionage, “Tracking GhostNet” the networks are quite separate.)

While responsibility for such attacks are often attributed to state entities, 2010 also saw a series of attacks linked to the Zeus malware that appeared a lot more like espionage than crime. After Netwitness released a report on the Kneber botnet, a Zeus-based botnet with domain names registered to, I focused on the connections between that botnet and a series of attacks against .mil and .gov email addresses using social engineering techniques. Have criminals determined that there is a market for sensitive data? It sure seems that way to me.

Abusing the Cloud
In “Shadows in the Cloud: An investigation into cyber espionage 2.0” my co-authors and I analyzed the command and control infrastructure of a network that used a complex and tiered command and control infrastructure that leveraged Twitter, Google Groups, Blogspot, Baidu Blogs, and Yahoo! Mail in order to maintain persistent control over the compromised computers. Of course, such techniques are not new, in 2009, researchers found that Twitter, Jaiku, Tumblr, Google Groups, Google AppEngine and Facebook had all been used as the command and control structure for malware. In 2010, Sunbelt found a Twitter botnet creator and Trend Micro reports that the “Here You Have” worm used GMail accounts.

During my analysis of malware posted on the Contagio blog, I noticed that the malware used an encrypted connection to Gmail as a means of command and control. (It also used cloud storage at in order to have the compromised computers download additional malware components). As network defenses continue to include traffic analysis, I believe that we will continue to see a move toward using popular services, especially web mail as command and control elements. Unlike connections to well-known dynamic DNS services, connections to Gmail and other popular services do not necessarily stand out and are encrypted.

Big Money
Although there are interesting target malware attacks that appear to have political motives, money continues to be the driving force behind the bulk of malware encountered by most Internet users. Cybercrime is profitable. In 2010, the Information Warfare Monitor released a report that documented the inner workings of Koobface. Koobface is a notorious botnet that leverages social networking platforms to propagate. The operators of Koobface have been able to successfully monetize their operations. Through the use of pay-per-click and pay-per-install affiliate programs, Koobface was able to earn over US$2 million between June 2009 and June 2010 by forcing compromised computers to install fake antivirus software and engage in click fraud. (BlackHat SEO operators monetize their operations in a similar way., see here and here.)

However, more traditional heists based on stolen banking and credit card credentials continue thanks to malware such as Zeus and SpyEye. This year, law enforcement were able to arrest individuals that used the Zeus malware to steal $70 million dollars. Often, these operations recruit money mules and pack mules to relay stolen money and goods bought with stolen credit cards. This makes it difficult to apprehend those behind these operations.

RX-promotion: A Pharma Shop

More than 65% of spam consists of “pharmaceutical spam” sent through a variety of well known spam botnets such as Rustock and Cutwail. These spam messages use multiple shop brands and sell a variety of drugs, especially Viagra. These pills, sometime fake pills, are shipped to buyers from pharma manufacturers, often in India or China.

There are pharma campaigns that have been found to use thousands of domain names and fast flux DNS techniques which can effectively resist takedown efforts. However, pharma operations are not just centralized campaigns. Much like FakeAV, Pay-Per-Install, and Pay-Per-Click operations, pharma is also organized through affiliate networks. Affiliate networks allow centralized pharma operations to diversify with individual operators maintaining pharma websites and generating incoming traffic through spam or search engine optimization.

This post will focus on RX-promotion ( which has been linked to a variety of operations including the payment service ChronoPay. The FDA are aware of Rx-promotion and have sent a warning letter to them in October 2010. Rather than focus on those behind the operation, this post simply focuses on how the affiliate program works and Rx-promotion’s pharma brands including:

The Canadian Rx Drugs –
Meds Leader – Top Online Pharmacy Supplier –
Health-Refill –
Men Drugs Shop –
The US Drugs –
Canadian Online Meds –
Trusted Meds Online –
MedrugsPlus –
Internet Drugs Pedia –
The Canadian Rx Drugs –
Always Great –
RXED On Green –
StallionsRX –
Golden StethoScope –
Star Of Health –
RX Pharmacy Center –
Cheap Meds List –
Health Online Leader –
Drugs For Us –
Meds For Us –
Great RX Pharmacy –
World Of Drugs –
Number One Clinic –

It is actually quite simple to get started with a pharma affiliate operation and there are even guides that walk users through the process. After creating an account, one can download rx.tar.gz, a package that allows anyone to setup a pharma shop of their own. RX-promotion operates a number pharma brands that are available as themes after setting up the pharma shop. (*See screen shots at the bottom of this post).

Although the brand shops are operated by the affiliates, the shops are connected to the Rx-Promotion infrastructure which provides the backend for the prices, payment and support. The shops make HTTP connections to and receiving XML updates. Payments are handled through and customer support is available at

As orders are received, the affiliate earns money from Rx-Promotion and can “cash out” through a variety of services.

There is ongoing development of the shop code. Affiliates can easily update their installations through the administrative backend.

In one such update, Rx-Promotion left the “./svn/” directory from the subversion revision control system indicating that their source code and development resides at:

RX-promotion is behind many different pharma brands that are marketed using spam and search engine optimization techniques. RX-promotion provides the backend of the pharma operation while numerous affiliates promote its products in order to receive a portion of the profit generated.

Screenshots of the various pharma brand themes available is the Rx-Promotion shop code:

























Pack Mules: The Re-Shipping Fraud & Malware Connection

Malware toolkits are designed to steal information, such as bank account data, and provide cyber criminals with vast quantities of stolen credentials. Every day, credit card numbers stolen by malware such as Zeus and SpyEye are bought and sold in the underground economy. This has given rise to the recruitment of “pack mules.”

When using stolen credit card numbers to make purchases online, criminals do not provide their own identity or location information. Instead, criminals post advertisements on job search Web sites in order to lure “pack mules” to act as intermediaries in their criminal operations. These intermediaries receive merchandise on the criminal’s behalf and re-ship it to a location under the control of the criminals. This operation is known as “re-shipping fraud” and is similar to the ways in which some criminals recruit “money mules” to open bank accounts for transferring stolen funds.

Re-shipping is tightly intertwined with malware activity. This is demonstrated by the fact that the Web sites used to recruit pack mules are hosted on the same servers that host the command-and-control servers of Zeus botnets. I have been exploring (see Clustering Zeus Command and Control Servers Part 1 and Part 2) clusters of Zeus activity in an attempt to better understand the connections among the criminals behind different functions within the botnet ecosystem. I have found that although Zeus is a popular malware toolkit that any aspiring criminal can use to setup a botnet capable of stealing credit card and banking information, there is a cluster of malicious Zeus servers which indicate that there is a “core” of Zeus operations.

In this blog post, I analyze the pack mule recruiting Web site, “Sullivan and Myers,” ( and explore its links with Zeus botnets and the broader malware underground. This investigation indicates that these concentrations of malicious activities go beyond operating command-and-control servers and extracting banking information to other aspects of the criminal enterprise. This includes exploitation (through “exploit packs“) and the recruiting of pack and money mules.

Pack Mule Recruitment

In order to recruit pack mules, criminals setup Web sites that purport to belong to a legitimate shipping and receiving business, and post advertisements that link to the “business” on job search Web sites and forums. This can be seen in the case of Sullivan and Myers, a fake business created for the purpose of recruiting pack mules.

Sullivan and Myer’s job posting invites interested applicants to complete an online application form and submit a resume to Sullivan and Myer’s contact information (address, phone, and fax number) is also supplied. The application form, contact information, and the company’s Web site appear to have been designed to create a sense of legitimacy. Although there are some indicators that suggest the company may be fake, such as awkward language and occasional errors (using “Myers & Sullivan” instead of “Sullivan and Myers”), the overall presentation is passable. To some applicants, the company may appear to be legitimate.

After submitting a resume, applicants are given additional information about the position. The applicants are informed that they will be receiving packages which they are to re-package and send to the company’s “consumers.” The applicants are told that they can earn up to USD3000 per month.

Human Resource

Your documents has been verified and checked; you seem to be a suitable
candidate for Junior Packing Specialists’ position and we are glad, that you are
interested in this opening.

Following, you’ll find information about Sullivan & Myers and additional details
about Junior Packing Specialist position.

Sullivan & Myers (NASDAQ: SUM) is a well known printing and typography company
that offers wide variety of printing, publishing and general advertising
services. Company is based in US with headquarters in GA, Atlanta. If you want
to find out more about Sullivan & Myers, please visit our web site

This is a part-time job with a flexible schedule. Work time is not
limited, but to be successful you need to devote at least 10hrs per week to it,
though those who work up to 20hr/week have best results in the company.

This is a part-time job and it can be rendered at home, thus all but few

communications will be handled online, because of this job requirements include
acceptable level of computer literacy and Internet access. There is no entrance
or any other hidden fee. The company covers all the fees related to this

Junior packing specialist’s job is quite simple, currently Sullivan & Myers
provide a complex package of services for a network of a well-known consumer’s
electronics company, you will be receiving scheduled packages from them. The
parcels mostly consist of electronics and consumer goods with no oversized
deliveries. You shall receive a specialized packing paper from Sullivan & Myers
or its affiliates, part of it will be a decal paper, picturing different
advertisements from our client’s partner, some might only be protective wrapping
to provide additional security to fragile goods. Junior Packing specialist’s job
is simple, you need to repack each package & parcel and make sure that
consistence of package is fully operational or/and lacking visual defects and
forward it to consumers via USPS or FED EX. You might receive up to 10 packages
per week (during your trial period) thus as we already mentioned we require at
least 10hrs to be dedicated to this job.

To the successful applicants we offer a position on a trial period (30
business days, from the first actual assignment). This is the period when you
will be trained and shall receive 24/7 online and phone support, while earning
money. The evaluation of employees on a trial period is usually at least one
week before the end of their trial period. During the trial period, the
supervisor can recommend termination. At the end of the trial period, supervisor
makes his decision.

The trial period is paid $1390 USD per month. For every successful mail/parcel
forwarded you will receive $35, also you shall receive an additional bonus of
$15 per parcel that you send at the day of delivery, for example, if you have
received a parcel at 01.05.2010 and forwarded it at the same day, you shall
receive not $35 but $50 commission. Your total income, with the current volume
of clients, will be added up to $3000 USD per month. Your base salary, after
trial period, will go up to $1900 per month, plus $45 per parcel you forward.

You may ask for additional hours after trial period, or proceed full-time.
If you are interested in this job, please reply to this e-mail and our HR
managers will send you all required paperwork.

Next, applicants are sent a contract and are then instructed to send copies of identification and proof of residency for a background check to minimize fraud. This is an important step because if, at a later point, the applicant determines that the company is not legitimate and wants to quit, the criminals behind this operation could attempt identity theft or otherwise compromise the individual.

Human Resource

In this e-mail, you will find attached legal document specifically a labor
contract for Junior Packing Specialist position in Sullivan & Myers.

Make sure you read it carefully, familiarize yourself with all aspects of
the agreement and in case if you agree with the terms do the following:

1. Print out two (2) copies of the labor contract.
2. Sign both parts, you must sign it on the bottom of EVERY page,
plus at the end of the document.
3. Forward one part to Sullivan & Myers HR department at or fax it to 1-(678)-866-2530
4. Keep one signed copy for yourself.

The contract becomes valid from the moment of the reception of the
correctly filled copy of the contract. It should be noted that the validity
of the contract in the electronic form is identical to the contract signed
in personal presence of both parties.

In order to minimize fraudulent activities we have implemented strong
security policy, we are running mandatory background checks for every
successful candidate. Background check includes but is not limited to,
criminal, financial or personal records that are available publicly. In VERY
rare cases, Sullivan & Myers may enforce PI. As a part of our security
policy we ask you to make an electronic copy of your ID, driving license or
any other legal document that may verify your identity (any utility bill
will do, if your domicile is mentioned there) and send it attached with the
same e-mail or fax it to 1-(678)-866-2530.

You will receive additional information when your forwarded contract will
be examined and verified by our attorneys.

*NOTE: Requires manual signature.

After receiving the signed contract, the criminals confirm the mailing address of the new “employee.” At this point, the new employee will begin receiving packages of goods bought with stolen credit card information and forwarding these goods to the criminals behind the operation. When law enforcement tracks down the operation, they will be led to the address of the pack mule rather than the masterminds behind the operation.

The Malware Connection

Locating Sullivan and Myers within the malware ecosystem exposes the criminal connections of those behind the re-shipping fraud operation. The Web site is registered to the e-mail address and resolves to the IP address is linked to significant malicious activity.

The hosting history of firmly places the domain within concentrations of malicious activity. Currently, the Web site is hosted on a server with the IP address This server also hosts (, ( and and are Web sites that host malware, and appears to be another pack mule recruiting Web site. has been hosted on a number of servers that have hosted significant amounts of malicious activity in the last year. Currently, these servers are hosting domain names registered to known malicious e-mail addresses.


– –
– –
– –
– –


– –
– –
– –
– –


– –


– –
– –
– –
– –
– –
– –
– –
– –
– –
– –
– –

The domain names listed above resolve to IP addresses of servers that were previously used to host While some of the domain names have already been linked to malicious activity, some have not. However, they are associated with e-mail addresses that have been used to register malicious domain names in the past.

Using data from MalwareDomainList and ZeusTracker, we can see the extent to which domain names registered by are engaged in malicious behavior and linked through co-hosting to other malicious domain names. These malicious domain names have been active throughout 2010 and have been used to host exploit packs, such as Pheonix and Eleonore; downloaders, such as Oficla/Sasfis, Fake Antivirus, the RussKill DDoS tool and multiple versions of the Zeus Trojan; and associated drop zones and command-and-control servers. This e-mail address was also used to register, a command-and-control server for the Ambler botnet.

The most interesting connection within this cluster links the activity of domain names registered with to the Ambler botnet and to a cluster of malicious Zeus activity. The domain name ( – was an Ambler command-and-control server that was operated by the same set of actors that administered a cluster of Zeus command-and-control servers registered with a variety of well- known e-mail addresses, including,, and

The e-mail address was made infamous after Netwitness revealed the existence of a Zeus-based botnet associated with that email address that had compromised over 74,000 computers around the world. An association with the Kneber botnet indicates that those behind the operation have no shortage of stolen credit card numbers that could be used to make purchases that are re-shipped through the pack mule operation. Moreover, this cluster was found to be not only operating a Zeus botnet, but a SpyEye and the Ambler botnet as well. This indicates that the criminals are diversifying their operations using multiple forms of malware that are designed to steal credit card numbers, bank account information, and other credentials.

However, there are some limitations to this analysis. Just because domain names are hosted on the same server, it does not mean that there is necessarily a direct connection between them. There are a variety of “bullet proof” Web hosting companies that provide stable hosting to a wide variety of malicious activity. Online criminal prefer these services because the “bullet proof” hosts ensure that malicious Web sites remain online despite efforts of the security community to take them down.

Domain names registered with the same e-mail address provides a stronger link because this indicates that the domain names are under the control of one entity. However, domain names registered to the same e-mail address may not be directly linked. There are a variety of services available within the malware underground that include domain registration. For example, the domain name ( is hosted on a server that was formerly hosted on. The server is also associated with a service that provides domain name registration. If domain registration services register domain names for multiple clients with the same e-mail address, it provides a weak (rather than strong) link between malicious activity clustered around domain names registered with the same e-mail address. Domain names registered with the same e-mail address may be distributed by the supplier to an array of disparate criminals. So, rather than indicating a strong connection between the malicious actors using the domain names, it simply shows that disparate malicious actors sought the services of the same domain name provider.

Keeping these limitations in mind, I believe that while there are specialized roles within the malware ecosystem, there appears to be a significant portion that is quite centralized. In this case, domain names registered with the same e-mail addresses not only inhabit servers full of malicious activity, but are also associated with “pack mule” recruitment, exploit packs, and Zeus and Ambler command-and-control servers. While the exact nature of the connections between them are unclear, these concentrations indicate that a discrete set of criminals are behind an operation that goes full circle—from exploiting victims, to harvesting credentials to acquire goods which are relayed through a network of pack mules back to the criminals.

Koobface: Inside a Crimeware Network

The Information Warfare Monitor (Citizen Lab, Munk School of Global Affairs, University of Toronto and the SecDev Group, Ottawa) announce the release of Koobface: Inside a Crimeware Network by Nart Villeneuve, with a foreword by Ron Deibert and Rafal Rohozinski.

The full report can be accessed here (local mirror):

Globe and Mail coverage of the report can be accessed here:

Koobface is a notorious botnet that leverages social networking platforms to propagate. Since, people are much more likely to execute a malicious file if it has been sent to them by someone they know and trust, the Koobface operators, known as “Ali Baba and 40 LLC” have developed a system that that uses social networking platforms such as Facebook to send messages containing malicious links. These links redirect users to false YouTube pages that encourage users to download malicious software masquerading as a video codec or a software upgrade.

In late April 2010, I discovered archive files on a well known Koobface servers that provided an inside look at the operations and monetization strategies of the Koobface botnet. The contents of these archives revealed the malware, code, and database used to maintain Koobface. It also revealed information about Koobface’s affiliate programs and monetization strategies. There are three main issues that have stood out for me throughout this investigation.

The first is the level of Koobface’s financial success. The operators of Koobface have been able to successfully monetize their operations. Through the use of pay-per-click and pay-per-install affiliate programs, Koobface was able to earn over US$2 million between June 2009 and June 2010 by forcing compromised computers to install malicious software and engage in click fraud. This, of course, does not occur in a vacuuum but within a malware ecosystem that sustains and monetizes botnet operations.

The second concerns the countermeasures taken by Koobface against the security community.Koobface maintains a banlist of IP addresses that are forbidden from accessing Koobface servers. In addition, Koobface operators carefully monitor whether any of their URLs have been flagged as malicious by or Facebook and they also monitor their malware links with the Google Safe Browsing API. This is part of a trend where malware authors check their malicious software against a variety of security products to ensure that there is only limited protection.

Finally, botnets such as Koobface present significant, but not impossible, challenges for law enforcement. Botnet operators leverage geography to their advantage, often exploiting Internet users from all countries but their own. While the total amount of criminal activity that the botnet operators engage in may be significant, the distribution of that criminal activity across multiple jurisdictions means that the criminal activity in any one jurisdiction is minimal. In addition, botnet operators leverage Internet infrastructure around the world, making it difficult to interfere with their operations.

However, botnet operators, such as those behind Koobface, do make mistakes. Information sharing and persistent monitoring can uncover the details of botnet operations. Therefore, it is important that the law enforcement and security community continue to share information and work closely together. An understanding of the inner workings of crimeware networks allows law enforcement to pursue leads and the security community to develop better defenses against malware attacks.

This report was made possible thanks to the guidance and encouragement of Ron Deibert and Rafal Rohozinski, the principal investigators of the Information Warfare Monitor. This report is built upon the research of members of the security community and I would like to thank all those who have documented the operations of Koobface over the years, especially Dancho Danchev and Trend Micro’s Threat Research Team. I would like to acknowledge and thank Chris Davis and Jose Nazario for sharing their knowledge and providing advice. In addition, I would like to thank the RCMP, the FBI, the UK Police, and AusCERT for their assistance. Finally, a special thanks is due to Jan Droemer who discovered the same data and shared his analysis and insights.

For more information on Koobface, see:

The Real Face of KOOBFACE: The Largest Web 2.0 Botnet Explained

“The Heart of KOOBFACE: C&C and Social Network Propagation

Show Me the Money! The Monetization of KOOBFACE

Web 2.0 Botnet Evolution: KOOBFACE Revisited

Koobface Gang Responds to the “10 Things You Didn’t Know About the Koobface Gang Post”

Koobface – the social network trojan

Nobel Peace Prize, Amnesty HK and Malware

There have been two recent attacks involving human rights and malware. First, on November 7, 2010, posted an analysis of a malware attack that masqueraded as an invitation to attend an event put on by the Oslo Freedom Forum for Nobel Peace Prize winner Liu Xiaobo. The malware exploited a known vulnerability (CVE-2010-2883) in Adobe Reader/Acrobat. The Committee to Protect Journalists was hit by the same attack.

On November 10, 2010 Websense reported that website of Amnesty Hong Kong was compromised and was delivering an Internet Explorer 0day exploit (CVE-2010-3962) to visitors. In addition, Websense reports that the same malicious server was serving three additional exploits: a Flash exploit (CVE-2010-2884), a QuickTime exploit (CVE-2010-1799) and a Shockwave exploit (CVE-2010-3653).

The malicious domain name hosting the exploits ( has been serving malware since Sept. 2010. The domain was registered in May 2010 to was formerly hosted on which now hosts the Zhejiang University Alumni Association website.

The malware dropped from the Internet Explorer exploit (CVE-2010-3962)
MD5: ca80564d93fbe6327ba6b094ae3c0445 VT: 2 /43

The malware dropped from the Flash exploit (CVE-2010-2884)
MD5: 0da04df8166e2c492e444e88ab052e9c VT: 2 /43

The malware dropped from the QuickTime exploit (CVE-2010-1799)
MD5: 3e54f1d3d56d3dbbfe6554547a99e97e VT: 16 /43

The malware dropped from the Shockwave exploit (CVE-2010-3653)
MD5: 3a459ff98f070828059e415047e8d58c VT: 0/43

Both ca80564d93fbe6327ba6b094ae3c0445 and 3a459ff98f070828059e415047e8d58c perform a DNS lookup for, which is an alias for which resolves to (China Unicom Beijing province network).

The domain name “” has been associated with a variety of malware going back to May 2010. This domain name, is registered to, the developer of the NetThief RAT.

Malware attacks leveraging human rights issues are not new. I have been documenting them for some time (see, Human Rights and Malware Attacks, Targeted Malware Attack on Foreign Correspondent’s based in China, “0day”: Civil Society and Cyber Security). However, one of the issues that Greg Walton and I raised last year, is a trend toward using the real web sites of human rights organizations compromised and as vehicles to deliver 0day exploits to the visitors of the sites – many of whom may be staff and supporters of the specific organization. Unfortunately, we can expect this to continue.

Clustering Zeus Command and Control Servers Part 2

In Part 1 of “Clustering Zeus Command and Control Servers” I focused on clustering Zeus command and control servers based on three criteria: IP addresses, domain names, and email addresses used to register domain names. Using data drawn from ZeusTracker and MalwareDomainList, I observed that while a wide variety of criminals may set up disparate Zeus operations there may be “core” set of Zeus operations clustered around domain names registered five email addresses:,,, and Beyond the common email addresses and co-hosting on servers with the same IP addresses (which, in general are hosting a wide variety of malware) the exact nature of the relationships remains unclear.

It is clear that there are certain servers that facilitate an abundance of malicious activity. However, caution must be exercised when conclusions are drawn regarding specific (groups of) actors operating discrete segments of botnet command and control servers among a common malicious infrastructure. Malware groups are often the customers of other malware groups or work with affiliates to propagate and monetize malware. Different groups may propagate malicious domain names that belong to other groups, or different groups may propagate common malicious domains that are provided by an affiliate network. In addition, there are malicious networks that provide hosting services to malware distributors and botnet operators. Therefore, links that appear between a variety of actors may not be as solid as the technical data alone would lead one to believe.

In order to examine these relationships further, I’m going to layer some qualitative data and analysis on the Zeus data analyzed in Part 1. Based on information I obtained from some of the command and control servers listed below (this is deliberately vague), combined with common file paths and the presence of the same files on different combinations of these servers, I believe that the following command and control domain names constitute of cluster of malicious activity operated by the same set of operators: – – – – – – – – – – – – – – – – – – – – – Steven Lucas – – – – – – – – – – –

This post will explore the relationships between these domains and other malicious activity, primarily Zeus activity, undertaken by other domain names registered with the same email addresses in order to explore the theory that there is a “core” of Zeus activity. While the malicious activity primarily relates to Zeus there are some significant exceptions. The domain name was used as a command and control server for the Ambler botnet. For the period I observed the Ambler activity, over 5000 IP addresses from compromised computers, 99% of which were from Russia, checked in with the command and control server. In addition, I found that was acting as a SpyEye command and control server in addition to a Zeus command and control server.

This screenshot shows the relationship between the command and control domain names, the malicious activity associated with them and the IP address that the domain name resolves to. While there are several instances in which some domain names were co-hosted on the same server, nearly half were not. This makes sense as operators will seek to diversify their hosting in order to avoid a complete shutdown should one of their command and control servers be taken down or blocked. In fact, look at the time span, covering October 2009 to September 2010 we can see how the operators moved their operations from one server to the next.

This operators of this malware cluster tend to host their command and control servers in Eastern Europe and China.

In order to assess this clusters possible linkages within the broader malware ecosystem, the data set was expanded to include a) other domain names registered with the same email addresses and b) the IP addresses of the servers associated with the malicious activity imported from ZeusTracker and MalwareDomainList. This extends the geographic scope of the hosting servers into North America, as well as the previous locations in Eastern Europe (UA, RU, CZ, MD) and South East Asia (CN, TW).

Looking at the relationships between the domains we see that there are two interesting clusters, and arguable a few smaller ones as well. These represent concentrations of servers registered with the same email addresses. The two main clusters are domain names registered to: and

An interesting fact about the “Lucas” cluster becomes apparent when you look at the time line of malicious activity (the date when the domain name was added to ZeusTracker or MalwareDomainList). The Lucas cluster is primarily active January – November 2009 (although there is some subsequent activity) while very few domains registered with other email addresses are active.

This is followed by the introduction of the “Kneber” domains which begin on the tail end of the Lucas cluster’s activity. The Kneber domain names begin in November 2009 and continue into October 2010. While the domain names registered with the remaining email addresses do also roughly follow a similar pattern of beginning while the previous one tails off, Kneber remains fairly constant once it begins.

In Part 1, I showed that there are clusters of Zeus activity that around a set of email addresses used to register domain names. Using qualitative data from my investigations, I’ve found a Zeus cluster that uses domain names registered by some, but not all, of these key email addresses including and This cluster has transitioned through domain names registered by a variety of email addresses over the last year. When the data set is expanded to include all the domain names registered by these email addresses in ZeusTracker and MalwareDomainList we see the same pattern of transition play out. This supports the theory that while Zeus is a toolkit that allows anyone to create a botnet, there is a “core” of Zeus activity.

However, this cluster of 16 domain names is only a small portion of the “core” Zeus activity associated with five key email addresses. According to DomainTools, about 1839 domain names in total: is associated with about 717 domains is associated with about 449 domains is associated with about 110 domains is associated with about 263 domains is associated with about 300 domains

These email addresses have been used to registered a variety of domain names associated with all manner of malicious activity, not exclusively Zeus activity. While this could be part of a centralized effort to distribute command and control servers to be operated by sub-groups, I am not sure that it is best to attribute all the malicious activity across these domains to the same set of actors. Even if these domain names represent the efforts of the same set of actors, they appear to be distributed to smaller groups of operators. These operators don’t necessarily have connections with others managing domain names hosted on the same infrastructure and/or registered with the same email addresses.

However, this simple clustering method does provide us with concentrations of malicious activity that should be investigated further. The introduction of qualitative data provides the ability to probe the operations of specific groups further. In the future I’d like to acquire a list of all 1800 domain names and layer on historical hosting data to see if any further patterns emerge.

Command and Control in the Cloud

In “Shadows in the Cloud: An investigation into cyber espionage 2.0” my co-authors and I analyzed the command and control infrastructure of a network that extracted secret, confidential and restricted documents from the Indian government and military. The Shadow Network used a complex and tiered command and control infrastructure that leveraged Twitter, Google Groups, Blogspot, Baidu Blogs, and Yahoo! Mail in order to maintain persistent control over the compromised computers. As we noted in the report, the use of these services as elements of command and control is certainly not new:

The use of social networking sites as elements of command and control for malware networks is not novel. The attackers leverage the normal operation of these systems in order to maintain control over compromised system. In 2009, researchers found that Twitter, Jaiku, Tumblr, Google Groups, Google AppEngine and Facebook had all been used as the command and control structure for malware. In August 2009, Arbor Networks’ Jose Nazario found that Twitter was being used as a command and control component for a malware network. In this case, the malware was an information stealer focused on extracting banking credentials from compromised computers located mostly in Brazil. Twitter was not the only channel being used by the attackers. They also used accounts on Jaiku and Tumblr (Nazario 2009a). Furthermore, Arbor Networks found another instance of malware that used the Google AppEngine to deliver malicious URLs to compromised computers (Nazario 2009b). The Unmask Parasites blog found that obfuscated scripts embedded in compromised web sites used the Twitter API to obscure their activities. While the method was clever, the code was unreliable and appeared to have been abandoned by the attackers (Unmask Parasites 2009). Symantec found that Google Groups were being used as command and control for another instance of malware. In this case, a private Google group was used by the attackers to send commands to compromised computers which then uploaded their responses to the same Group (Symantec 2009a) Symantec also found an instance of malware that used Facebook status messages as a mechanism of command and control. (Symantec 2009b). The use of these social networking and Web 2.0 tools allows the attackers to leverage the normal operation of these tools to obscure the command and control functions of malware.

Earlier this year, Sunbelt found a Twitter botnet creator and Trend Micro reports that the “Here You Have” worm used GMail accounts. As we found with the Shadow Network malware authors learn from each other. And in the case of the Shadow Network they didn’t just use one service they used six of them, including Yahoo! Mail. And while indiscriminate malware may be rather noisy, the malware used in targeted attacks tends to be (but is certainly not always) more discrete.

A recent sample posted at caught my attention for this very reason. The sample, “Conference Information_2010 IFANS Conference on Global Affairs (1001).pdf” (which was sent from and was detected by 14 /43 (32.6%) AV products at Virustotal) arrived with the subject line “Nuclear Challenges and Responses in the Century” and exploited a vulnerability in Adobe Reader/Acrobat (CVE-2010-2883) to drop malware on the targets’ computers. For those of you who follow Mila’s awesome blog, this scenario is hardly surprising.

But a few things caught my attention. There were references in the strings dumped from a file the malware created (syschk.ocx) that referenced GMail ( and DriveHQ (, which describes itself as a “cloud based storage, backup, group sharing and collaboration service.” When you look at the traffic generated by the malware you’ll see connections to these locations.

There is nothing about these locations that is very suspicious — everyone checks their GMail right? Moreover, the connection to GMail is SSL encrypted.

Using Burp (which made the process very simple) I MITM’d the traffic between the malware and GMail. The malware logs in to the GMail account and sends an email to another GMail address. The content of this email is encrypted. However, I believe that what it is sending — although this is just a hunch — is the content of another file the malware generates: form.ocx. This file contains what appears to be a unique ID assigned by the malware, the hostname and IP address, the default home page of the default browser and a listing of installed programs on the computer. The end of the file contains information about executables the malware has impacted. In addition to the encrypted message sent through the GMail account, the Unique ID in form.ocx appears at the beginning of the message.


Infect OK!

I have not looked into what exactly the malware does to these applications, but it basically disables the operation of FireFox and Chrome and instead connect to the Gmail account when you try to start these applications. Internet Explorer seems to function normally.

The connection to results in the download of an additional file rename.ocx which appears to be very similar, when its strings are compared with, syschk.ocx. It then renames syschk.ocx to syschk.ocx1. You can see that this correlates with text in the strings dumped from syschk.ocx.


After the initial connections to GMail and DriveHQ the malware went quiet. I never did get it to connect again.

As network defenses continue to include traffic analysis, I believe that we will continue to see a move toward using popular services, especially web mail as command and control elements. Unlike connections to well-known dynamic DNS services like or abnormal connections to geographic regions, connections to GMail and other popular services do not necessarily stand out. Moreover, the connections to the services, such as GMail are encrypted, further obfuscating the malicious activity that is occurring.

Malware Diversification

There are wide varieties of malware, many of which have similar functionality. As a result there is a tendency to portray them as being in competition with on another. In some ways this is true, especially when it comes to malware authors, however, I prefer to see it as less of a rivalry and more of an opportunity for diversification on the part of the botnet operators. Recently there have been some articles that suggest that Zeus may be “dethroned” (“New threat set to dethrone Zeus“, “Online criminals are moving on from Zeus“) thanks to Bugat and Carberp.

Well, despite the recent arrests of over 150 individuals associated with Zeus-related bank fraud and the decline in the number of active Zeus command and control servers Zeus is still “going strong” and demonstrating its resilience.

However, this is not a property of the malware, but of the wide base of criminals that use it. While there may be a core of Zeus activity, anyone can use the Zeus toolkit to setup his or her own botnet. An additional factor to include is the fact that criminals make use of multiple malware kits, even rival malware kits.

The relationship between SpyEye (see two great SpyEye posts here and here) and ZeuS has been described as a rivalry — largely based on SpyEye’s ability to remove ZeuS from compromised computers — but botnet operators make use of both.

Here are two command and control server domain names that have hosted both Zeus and SpyEye. The domain was a known Zeus (see MDL) command and control, but I found that it was also hosting SpyEye. More recently, I have been monitoring that was a known Zeus command and control (see MDL) but is also hosting SpyEye.
b911f40ff9573f33e73055b2267a5cd7 bd.exe
VT: 36/ 43 (83.7%)
e8091d2099a8472b27a62c5ae57be5e9 id.exe
VT: 37/ 43 (86.0%)

Malware diversification allows the botnet operators to run multiple botnets, increasing their resilience to countermeasures aimed at taking down one particular strain. In addition, they can capitalize on new features and functionality available across various toolkits. To counter such operations we need to look beyond the toolkit and and investigate the operators as well.

Clustering Zeus Command and Control Servers

Recently, more than 150 individuals around the world have been arrested on bank fraud related charges after using the Zeus malware to acquire credentials that enabled the criminals to steal more than $70 million dollars. Those arrested include five Ukrainian individuals that are believed to be the masterminds behind the operation. Brian Krebs notes that there is a correlation between the decreasing number of active Zeus command and control servers and the timing of the arrests.

This is interesting because while “the media” often portrays Zeus as “a botnet” the security community rightly points out that Zeus is a malware toolkit not “a” botnet and that there are multiple Zeus botnets. However, what explains the decrease in Zeus command and control servers with the disruption of just one Zeus operation? While it is certainly true that any aspiring criminal can acquire Zeus and begin his or her own operation, is there a Zeus “core” that is organized and connected through links the criminal underground? Having just returned from Palantir’s Govcon feeling inspired I imported Zeus data from the MalwareDomainList and the ZeusTracker to explore the links between Zeus command and control servers.

While there are definitely more indicators, I focused on three: IP addresses, domain names, and email addresses used to register domain names. The IP addresses represent the servers that are used to host command and control servers. One such server may host multiple command and control servers allowing one to cluster malicious domain names that are hosted on the same server. Domain names are useful indicators but essential have a one-to-one relationship so it is more valuable to cluster them by the email address used to register the domain name. Using these indicators the Zeus command and control domain names can be clustered based on co-hosting (on the same IP address) and mutual registration (same email address). This may provide some indication if there is a “core” or Zeus activity.

However, there are significant limitations to bear in mind. Malicious hosting services are available in the criminal underground, so while a single server may be a hotspot of malware activity, it may not be directly related. On the other hand, some command and control servers may be using fast flux which would negate clustering by IP address altogether. Some command and control servers are based on IP addresses only and do not have domain names associated with them. On the other hand, a single domain name may be used for a variety of purposes. (For example, I have found a domain name that hosts both a Zeus and a SpyEye command and control server, despite the reported rivalry between them). In addition, the botnet operators may register a variety of domain names from a variety of email addresses. In such cases, clustering by email addresses would not yield significant links. Finally, there may be suppliers of domain names in them malware underground that register domain names with email addresses under their control, but sell the domains names to other criminals. In such cases, while the email address may be the same, the operators of botnets may not be directly related.

The data set used contains 5,907 domain names (control servers) and 4,505 IP addresses (servers) drawn from ZeusTracker and MalwareDomainList (where the activity on MDL contains “zeus”). Here, 4,505 IP addresses have been geocoded (not all were successfully geocoded) and displayed using Palantir’s heatmap. While there is Zeus activity hosted all over the world, there are noticeable concentrations in Europe, the Unites States and China.

This cluster on the Palantir graph represents the relationship between 5,907 domain names (control servers) and 4,505 IP addresses (servers). This initial display highlights a few interesting indicators. There are several clusters that are visually apparent which show multiple domain names hosted on one server (there are three prominent “star” clusters and several smaller ones) and there is a discernible “tree” structure in the center indicating relationships between single domain names that have been hosted on multiple IP addresses. And we can see thaht there are some familiar IP addresses used to register multiple domain names, the most notable being “” which is the email addresses behind the Kneber botnet.

Zooming in to some of the clusters reveals some interesting behaviors. In this example, one server is hosting 60 domain names. These 60 domain names were registered with 17 different email addresses. And when some additional information from MDL is brought in, we see that most of the domains are hosting a Zeus executable with the same name “patch.exe” and that there is a naming convention. For example, “” was registered with “” while “” was registered with “”. These domain names were all added to MDL around the same time and despite the multiple email addresses it does appear as if this is a single campaign.

In order to explore the question of whether or not there is a Zeus “core” of some sort, I filtered the domain names and IP addresses to those registered with the top five appearing email addresses (with the exception of which is the email address given for those who have used this domain privacy service). Domain names registered with these five email addresses account for 6.09% (360/5907) of the Zeus command and control servers. However, this number increases to 17.9% (360/2004) when the number of control servers is restricted to those that contain email data. In addition to several “star” clusters as well a “tree” in the middle of the graph, we see that these email addresses have been actively propagating Zeus for approximately one year. (The time is derived from when the domain is added to either the MDL or ZeusTracker lists, which is used a rough indicator of when a domain became active).

When the selection is restricted to only those domain names registered by “” we can see that these domains are represented across most of the clusters indicating that many of these domain are co-hosted on the same IP addresses with those registered by our other top email addresses. In addition, the “kneber” domain names are active through this year long period of data.

While a wide variety of criminals may set up disparate Zeus operations, clustering the Zeus command and control infrastructure in this way indicates that there is some evidence to support claims of a “core” set of Zeus operations. This may be one explanation for the observed decrease in active Zeus command and control servers.

However, this data only reflects only the relationships between IP addresses, domain names and the email addresses used to register the domain names. There are a variety of additional factors, especially those related to analysis of Zeus malware binaries that may support these linkages, provide additional linkages or challenge these linkages. Historical data showing coordinated movements to new IP addresses and name servers would provide additional means to cluster command and control servers with a higher degree of accuracy.

In Part 2 of this post I will broaden the analysis in order to see if the tentative conclusion hold with the introduction of additional data.

Black Hat SEO, PPC & RogueAV Part 2

Part 1 of “Black Hat SEO, PPC & RogueAV” focused on the type and amount of incoming traffic generated through BlackHat SEO methods. This traffic is monetized through the use of RogueAV, Pay-Per-Click and Pay-Per-Install affiliates. This post continues the analysis of this campaign by providing a inside look at this BHSEO operation.

The attackers acquired lists of thousands of FTP server credentials. The attackers may have purchased the compromised accounts from others in the cybercrime underground or harvested them from other operations. The attackers use several scripts to login to the FTP servers and upload their SEO scripts. The initial script uploaded to the compromised servers performs the following functions:

  • downloads the latest version of a redirection script
  • downloads a list of search queries
  • creates the files “tpl.txt”, “folders.txt” and “.htaccess”
  • creates a directory “wp-blog” that contains the files “go.php” (the downloaded redirection script), “keys.txt”, “nishe.txt”, “pages.txt” and “.htaccess”

The list of search queries are paired with random file paths in order to create pages on demand based on the search queries. When a request comes in, the redirection script check to see if the “referer” is from a search engine and if the the request appears to have been made by a “bot”. The latter function is performed by parsing the “user agent” header to check, for example, for indicators of a search engine crawler. If the “referer” is a search engine and the request is not made by a “bot”, the request is redirected to the SEO server. If either of these checks fail, the script will lookup the requested path to retrieve the search query it has been paired with.||chakra labels printable||t mobile rebate printable||printable instructions for sand castles||printable hanukkah song lyrics||4tth of july printable crown

Then the script will take the search query and retrieve the results for the query from Google and display the content using the “tpl.txt” file, which is a template based on the look and feel of the compromised website. The links in the page point to the additional search query / file path pairings.

These pages are indexed by search engines and the search queries become associated with the malicious pages. In addition, when a user queries a search engine, and lands on the malicious page, the user’s request is redirected to the SEO server along with the query that the user searched for. These queries are collected and feed into the search query lists used by the attackers.

At last count the attackers had uploaded their SEO scripts to 11,978 servers, and although the server appears to have been abandoned on 2010-09-20 the figures from earlier in the campaign indicate that the attackers were able to attract significant amounts of traffic.

The attackers recorded the referring domain name as well as the search query used to arrive at the compromised domain. These records along with the number of hist were recorded by the attackers and available from an unprotected web interface.

In order to monetize their operation, the attackers used several affiliates. Users that the attackers detected were running non-Windows operating systems were redirected to pay-per-click affiliates at these domain names: and Windows users were redirected to RogueAV landing pages.

The Rogue AV affiliates supply “landing page” URLs to their fake scanning pages that attempt to trick the user into installing the fake security software. These URL’s change over time, and the attackers maintain scripts that update these URLs so that user are redirected to fresh URLs that are less likely to have been identified and blocked by the security community.


file: db2d504abeedce8b404a1f5514989689 powersecure_2049_emr7.exe
VT: 3 /43 (7.0%)


url 1:[…]
url 2:[…]
file: 90245bf674ff3b16653fc6f7d191dead packupdate107_289.exe
VT: 18 /43 (41.9%)

Pay-Per-Install (PPI)

file: 02e62d95997b7db323175910bf14e19c file.1.exe
VT: 10/ 43 (23.3%)

This affiliate provides a URL that produces dynamic malware binaries. The attackers attempt to trick users into installing the malware by pretending that it is Adobe’s Flash player. The attackers script periodically queries the affiliate’s distribution point to receive a new binary, each new binary has a different hash value.

In addition, the attackers used malware detection services to scan the binaries to see how AV products detected them. The attackers used, which Brian Krebs documented earlier this year, as well as

When executed this trojan attempt to connect to and along with several other domains (,,,,, followed by numerous connections to ad servers.

In summary, this is not a complicated operation and is largely automated. The system collects what users search for and then creates fake pages based on those queries. search engines are fed these bogus pages and users are redirected to the SEO server that collects statistical information and the forward the user on to a monetization strategy either RogueAV, PPI or PPC. All the attackers need is a fresh supply of compromised FTP credentials which can be purchased in the cybercrime underground.


The Kraja botnet has managed to compromise 185,645 computers, the vast majority of which are located in Russia. Of the 199,513 unique IP addresses recorded from compromised computers, 87.88% are in IP address ranges assigned to Russia. The name “Kraja botnet” comes from an image located on the command and control server which was originally discovered by The Kraja botnet is related to the “Shiz” malware that was recently documented by Arbor Networks. Arbor concluded that this malware is also related to the “rohimafo” family. One of the more interesting observations made by Arbor is that the “shiz” malware will attempt to null-route specific IP address ranges that include a variety of security companies such as F-SECURE, KASPERSKY, SOPHOS, SYMANTEC, MCAFEE and TREND MICRO as well as various sandbox and analysis tools.

My analysis of the command and control infrastructure reveals that between Sept 11, 2010 to Sept 27, 2010 there were 185,645 compromised computers (using 199,513 unique IP addresses) that requested one of the following PHP files on the command and control servers: knock.php, knok.php and socks.php. In addition to recording the IP addresses of the compromised machines, I was able to record what appears to be a unique ID number for each compromised computer. The vast majority of the compromised computers requested knock.php.

knock.php – sends a configuration file to the compromised machine with what appear to be alternative command and control locations as well as the “magic” URL which appears to be a monetization strategy based on pay-per-click.

knok.php – sends the name of the operating system to the control server (may determine which computers to install a socks proxy on).

socks.php – sends the SOCKS proxy port number to control server.

While there was geographic distribution, a total of 87.88 percent of the IP addresses of the compromised hosts are in ranges assigned to Russia. This is significant because Russian and CIS are often not targeted by botnets because most of the PPC/PPI affiliates do not pay for clicks and installs from Russian IP addresses. It is also significant because Russian law enforcement generally requires there to be Russian victims in order to proceed with investigations. In this case, Russian appear to be the target of this botnet.

Old Threats are Current Threats

Despite the fact that the authors of the Pinch Trojan were “pinched” by law enforcement in 2007, the Pinch Trojan continues to be a current threat both because the source code is available (so anyone can modify it and release a variant) but also because old versions of Pinch continue to be effectively used. In 2007, F-Secure analyzed data collected from a Pinch command and control server using a tool called PinchParserPro PinchParserPro allows the attackers to parse, search and export the data stolen by the Pinch Trojan. Three years later Pinch is still in action, often bundled with an assortment of other malware. (Here is a paper that has a detailed technical analysis of Pinch variants.)

Data recovered from a recently active Pinch command and control server, (formerly indicates that 26,308 IP addresses uploaded data to the server. The top three countries affected were the US, Germany and Turkey but there was a considerable geographic distribution with a total of 150 countries affected.

In order to read the data PinchParserPro had to be used, which is an older than version than what F-Secure used (PinchParserPro in 2007. It is interesting that such an old version is still being successfully deployed.

While investigating the recovered data, credentials associated with government accounts were discovered. One of the victims of the malware was the Ministry of Foreign Affairs of the People’s Republic of China. While there has been much attention on malware attacks emanating from China, China is also a victim of malware attacks. In fact, a recent cyber-crime report by Symantec revealed that Chinese users were the most victimized by online crime.

The governmental accounts recovered from the control server include:

  • Ministry of Foreign Affairs, China
  • Industrial and Commercial Administration Bureau in Taiyuan, China
  • Ministry of Health, Turkey
  • Izmir Tax Services Department, Turkey
  • Istanbul Security Directorate, Turkey
  • Aegean Obstetrics and Gynecology Training and Research Hospital, Turkey
  • Ministry of Environment, Brazil
  • Regional Labor Court 6th Region, Brazil
  • National Electoral Commision, Poland
  • Ministry of Agriculture, Forestry and Water Management, Macedonia
  • Drug Enforcement Administration, Office of Diversion Control, E-Commerce program, USA
  • City of Oklahoma City, USA
  • Taipei Sewage Systems Office of Health, Taiwan
  • Ministry of Interior, Ukraine
  • Dirección Nacional de los Registros Nacionales, Argentina

While there is often an emphasis on the latest malware threats, old malware persists and continues to be very effective. In addition, attackers are able to compromise government systems using these outdated tools. And, even if the attackers did not intend to compromise these system — and I don’t think they did — attackers are, in general, beginning to realize that not all compromises are the same and that there may be additional value that can be extracted from particular compromised machines.