Malware Alerts

Subscribe to Malware Alerts feed Malware Alerts
Online headquarters of Kaspersky Lab security experts.
Updated: 1 hour 47 min ago

Threat Predictions for Industrial Security in 2018

Wed, 11/15/2017 - 05:00

The landscape in 2017

2017 was one of the most intense in terms of incidents affecting the information security of industrial systems. Security researchers discovered and reported hundreds of new vulnerabilities, warned of new threat vectors in ICS and technological processes, provided data on accidental infections of industrial systems and detected targeted attacks (for example, Shamoon 2.0/StoneDrill). And, for the first time since Stuxnet, discovered a malicious toolset some call a ‘cyber-weapon’ targeting physical systems: CrashOverride/Industroyer.

However, the most significant threat to industrial systems in 2017 was encryption ransomware attacks. According to a Kaspersky Lab ICS CERT report, in the first half of the year experts discovered encryption ransomware belonging to 33 different families. Numerous attacks were blocked, in 63 countries across the world. The WannaCry and ExPetr destructive ransomware attacks appear to have changed forever the attitude of industrial enterprises to the problem of protecting essential production systems.

What can we expect in 2018?
  1. A rise in general and accidental malware infections. With few exceptions, cybercriminal groups have not yet discovered simple and reliable schemes for monetizing attacks on industrial information systems. Accidental infections and incidents in industrial networks caused by ‘normal’ (general) malicious code aimed at a more traditional cybercriminal target such as the corporate networks, will continue in 2018. At the same time, we are likely to see such situations result in more severe consequences for industrial environments. The problem of regularly updating software in industrial systems in line with the corporate network remains unresolved, despite repeated warnings from the security community.
  2. Increased risk of targeted ransomware attacks. The WannaCry and ExPetr attacks taught both security experts and cybercriminals that operational technology (OT) systems are more vulnerable to attack than IT systems, and are often exposed to access through the Internet. Moreover, the damage caused by malware can exceed that in the corresponding corporate network, and ‘firefighting’ in the case of OT is much more difficult. Industrial companies have demonstrated how inefficient their organization and staff can be when it comes to cyberattacks on their OT infrastructure. All of these factors make industrial systems a desirable target for ransomware attacks.
  3. More incidents of industrial cyberespionage. The growing threat of organized ransomware attacks against industrial companies could trigger development of another, related area of cybercrime: the theft of industrial information systems data to be used afterwards for the preparation and implementation of targeted (including ransomware) attacks.
  4. New underground market activity focused on attack services and hacking tools. In recent years, we have seen growing demand on the black market for zero day exploits targeting ICS. This tells us that criminals are working on targeted attack campaigns. We expect to see this interest increase in 2018, stimulating the growth of the black markets and the appearance of new segments focused on ICS configuration data and ICS credentials stolen from industrial companies and, possibly, botnets with ‘industrial’ nodes offerings. Design and implementation of advanced cyberattacks targeting physical objects and systems requires an expert knowledge of ICS and relevant industries. Demand is expected to drive growth in areas such as ‘malware-as-a-service’, ‘attack-vector-design-as-a-service’, ‘attack-campaign-as-a-service’ and more.
  5. New types of malware and malicious tools. We will probably see new malware being used to target industrial networks and assets, with features including stealth and the ability to remain inactive in the IT network to avoid detection, only activating in less secure OT infrastructure. Another possibility is the appearance of ransomware targeting lower-level ICS devices and physical assets (pumps, power switches, etc.).
  6. Criminals will take advantage of ICS threat analyses published by security vendors. Researchers have done a good job finding and making public various attack vectors on industrial assets and infrastructures and analyzing the malicious toolsets found. However, this could also provide criminals with new opportunities. For example, the CrashOverride/Industroyer toolset disclosure could inspire hacktivists to run denial-of-service attacks on power and energy utilities; or criminals may targeted ransomware and may even invent monetizing schemes for blackouts. The PLC (programmable logic controller) worm concept could inspire criminals to create real world malicious worms; while others could try to implement malware using one of standard languages for programming PLCs. Criminals also could recreate the concept of infecting the PLC itself. Both these types of malware could remain undetected by existing security solutions.
  7. Changes in national regulation. In 2018, a number of different cybersecurity regulations for industrial systems will need to be implemented. For example, those with critical infrastructures and industrial assets facilities will be compelled to do more security assessments. This will definitely increase protection and awareness. Thanks to that, we will probably see some new vulnerabilities found and threats disclosed.
  8. Growing availability of, and investment in industrial cyber insurance. Industrial cyber-risk insurance is becoming an integral part of risk management for industrial enterprises. Previously, the risk of a cybersecurity incident was excluded from insurance contracts – just like the risk of a terrorist attack. But the situation is changing, with new initiatives introduced by both cybersecurity and insurance companies. In 2018, this will increase the number of audits/assessments and incident responses undertaken, raising cybersecurity awareness among the industrial facility’s leaders and operators.

APT Trends report Q3 2017

Tue, 11/14/2017 - 04:41


Beginning in the second quarter of 2017, Kaspersky’s Global Research and Analysis Team (GReAT) began publishing summaries of the quarter’s private threat intelligence reports in an effort to make the public aware of what research we have been conducting.  This report serves as the next installment, focusing on important reports produced during Q3 of 2017.

As stated last quarter, these reports will serve as a representative snapshot of what has been offered in greater detail in our private reports in order to highlight significant events and findings we feel most should be aware of. For brevity’s sake, we are choosing not to publish indicators associated with the reports highlighted. However, if you would like to learn more about our intelligence reports or request more information for a specific report, readers are encouraged to contact:

Chinese-Speaking Actors

The third quarter demonstrated to us that Chinese-speaking actors have not “disappeared” and are still very much active, conducting espionage against a wide range of countries and industry verticals.  In total, 10 of the 24 reports produced centered around activity attributed to multiple actors in this region.

The most interesting of these reports focused on two specific supply chain attacks; Netsarang / ShadowPad and CCleaner.  In July 2017, we discovered a previously unknown malware framework (ShadowPad) embedded inside the installation packages hosted on the Netsarang distribution site.   Netsarang is a popular server management software used throughout the world.  The ShadowPad framework contained a remotely activated backdoor which could be triggered by the threat actor through a specific value in a DNS TXT record.  Others in the research community have loosely attributed this attack to the threat actor Microsoft refers to as BARIUM.  Following up on this supply chain attack, another was reported initially by Cisco Talos in September involving CCleaner, a popular cleaner / optimization tool for PCs.  The actors responsible signed the malicious installation packages with a legitimate Piriform code signing certificate and pushed the malware between August and September.

Q3 also showed China is very interested in policies and negotiations involving Russia with other countries.  We reported on two separate campaigns demonstrating this interest.  To date, we have observed three separate incidents where Russia and another country hold talks and are targeted shortly thereafter, IndigoZebra being the first.  IronHusky was a campaign we first discovered in July targeting Russian and Mongolian government, aviation companies, and research institutes.  Earlier in April, both conducted talks related to modernizing the Mongolian air defenses with Russia’s help.  Shortly after these talks, the two countries were targeted with a Poison Ivy variant from a Chinese-speaking threat actor.  In June, India and Russia signed a much awaited agreement to expand a nuclear power plant in India, as well as further define the defense cooperation between the two countries.  Very soon after, both countries energy sector were targeted with a new piece of malware we refer to as “H2ODecomposition”.  In some case this malware was masquerading as a popular Indian antivirus solution (QuickHeal).  The name of the malware was derived from an initial RC5 string used in the encryption process (2H2O=2H2+O2) which describes a chemical reaction used in hydrogen fuel cells.

Other reports published in the third quarter under chinese-speaking actors were mainly updates to TTPs by known adversaries such as Spring Dragon, Ocean Lotus, Blue Termite, and Bald Knight.  The Spring Dragon report summarized the evolution of their malware to date.  Ocean Lotus was observed conducting watering hole attacks on the ASEAN website (as done previously) but with a new toolkit.  A new testing version of Emdivi was discovered in use by Blue Termite as well as their testing of CVE-2017-0199 for use.  Finally, Bald Knight (AKA – Tick) was seen using their popular XXMM malware family to target Japan and South Korea.

Below is a summary of report titles produced for the Chinese region.  As stated above, if you would like to learn more about our threat intelligence products or request more information on a specific report, please direct inquiries to

  1. Analysis and evolution of Spring Dragon tools
  2. EnergyMobster – Campaign targeting Russian-Indian energy project
  3. IronHusky – Intelligence of Russian-Mongolian military negotiations
  4. The Bald Knight Rises
  5. Massive watering holes campaign targeting Asia-Pacific
  6. Massive Watering Holes Campaign Targeting AsiaPacific – The Toolset
  7. NetSarang software backdoored in supply chain attack – early warning
  8. ShadowPad – popular server management software hit in supply chain attack
  9. New BlueTermite samples and potential new wave of attacks
  10. CCleaner backdoored – more supply chain attacks
Russian-Speaking Actors

The third quarter was a bit slower with respect to Russian speaking threat actors.  We produced four total reports, two of which focused on ATM malware, one on financial targeting in Ukraine and Russia, and finally a sort of wrap-up of Sofacy activity over the summer.

The ATM related reports centered around Russian speaking actors using two previously unknown pieces of malware designed specifically for certain models.  “Cutlet Maker” and “ATMProxy” both ultimately allowed the users to dispense cash at will from a chosen cartridge within the ATMs.  ATMProxy was interesting since it would sit dormant on an ATM until a card with a specific hard coded number was inserted, at which point it would dispense more cash than what was requested.

Another report discussed a new technique utilizing highly targeted watering holes to target financial entities in Ukraine and Russia with Buhtrap.  Buhtrap has been around since at least 2014, but this new wave of attacks was leveraging search engine optimization (SEO) to float malicious watering hole sites to the top of search results, thus providing more of a chance for valid targets to visit the malicious sites.

Finally, we produced a summary report on Sofacy’s summertime activity.  Nothing here was groundbreaking, but rather showed the group remained active with their payloads of choice; SPLM, GAMEFISH, and XTUNNEL.  Targeting also remained the same, focusing on European defense entities, Turkey, and former republics.

Below is a list of report titles for reference:

  1. ATMProxy – A new way to rob ATMs
  2. Cutlet maker – Newly identified ATM malware families sold on Darknet
  3. Summertime Sofacy – July 2017
  4. Buhtrap – New wave of attacks on financial targets
English-Speaking Actors

The last quarter also had us reporting on yet another member of the Lamberts family.  Red Lambert was discovered during our previous analysis of Grey Lambert and utilized hard coded SSL certificates in its command and control communications.  What was most interesting about the Red Lambert is that we discovered a possible operational security (OPSEC) failure on the actor’s part, leading us to a specific company who may have been responsible, in whole or in part, for the development of this Lambert malware.

  1. The Red Lambert
Korean-Speaking Actors

We were also able to produce two reports on Korean speaking actors, specifically involving Scarcruft and Bluenoroff.  Scarcruft was seen targeting high profile, political entities in South Korea using both destructive malware as well as malware designed more for espionage.  Bluenoroff, the financially motivated arm of Lazarus, targeted a Costa Rican casino using Manuscrypt.  Interestingly enough, this casino was compromised by Bluenoroff six months prior as well, indicating they potentially lost access and were attempting to get back in.

Report titles focusing on Korean-speaking actors:

  1. Scent of ScarCruft
  2. Bluenoroff hit Casino with Manuscrypt
Other Activity

Finally, we also wrote seven other reports on “uncategorized” actors in the third quarter.  Without going into detail on each of these reports, we will focus on two.  The first being a report on the Shadowbrokers’ June 2017 malware dump.  An anonymous “customer” who paid to get access to the dump of files posted the hashes of the files for the month, mainly due to their displeasure in what was provided for the money.  We were only able to verify one of nine file hashes, which ended up being an already known version of Triple Fantasy.

The other report we’d like to highlight (“Pisco Gone Sour”) is one involving an unknown actor targeting Chilean critical institutions with Veil , Meterpreter, and Powershell Empire.  We are constantly searching for new adversaries in our daily routine and this appears to be just that.  The use of publicly available tools makes it difficult to attribute this activity to a specific group, but our current assessment based on targeting is that the actor may be based somewhere in South America.

  1. Dark Cyrene – politically motivated campaign in the Middle East
  2. Pisco Gone Sour – Cyber Espionage Campaign Targeting Chile
  3. Crystal Finance Millennium website used to launch a new wave of attacks in Ukraine
  4. New Machete activity – August 2017
  5. ATMii
  6. Shadowbroker June 2017 Pack
  7. The Silence – new trojan attacking financial organizations
Final Thoughts

Normally we would end this report with some predictions for the next quarter, but as it will be the end of the year soon, we will be doing a separate predictions report for 2018.  Instead, we would like to point out one alarming trend we’ve observed over the last two quarters which is an increase in supply chain attacks.  Since Q2, there have been at least five incidents where actors have targeted the supply chain to accomplish their goals instead of going directly after the end target; MeDoc, Netsarang, CCleaner, Crystal Finance, and Elmedia.  While these incidents were not the result of just one group, it does show how the attention of many of the actors out there may be shifting in a direction that could be much more dangerous.  Successfully compromising the supply chain provides easy access to a much wider target base than available through traditional means such as spear phishing.  As an added benefit, these attacks can remain undetected for months, if not longer.  It remains to be seen if this trend will continue into 2018, but given the successes from the five mentioned above, we feel we haven’t seen the last of this type of attack in the near future.

IT threat evolution Q3 2017. Statistics

Fri, 11/10/2017 - 05:45

Q3 figures

According to KSN data, Kaspersky Lab solutions detected and repelled 277,646,376 malicious attacks from online resources located in 185 countries all over the world.

72,012,219 unique URLs were recognized as malicious by web antivirus components.

Attempted infections by malware that aims to steal money via online access to bank accounts were registered on 204,388 user computers.

Crypto ransomware attacks were blocked on 186283 computers of unique users.

Kaspersky Lab’s file antivirus detected a total of 198,228,428 unique malicious and potentially unwanted objects.

Kaspersky Lab mobile security products detected:

  • 1,598,196 malicious installation packages;
  • 19,748 mobile banking Trojans (installation packages);
  • 108,073 mobile ransomware Trojans (installation packages).
Mobile threats Q3 events The spread of the Asacub banker

In the third quarter, we continued to monitor the activity of the mobile banking Trojan Trojan-Banker.AndroidOS.Asacub that actively spread via SMS spam. Q3 saw cybercriminals carry out a major campaign to distribute the Trojan, resulting in a tripling of the number of users attacked. Asacub activity peaked in July, after which there was a decline in the number of attacks: in September we registered almost three times fewer attacked users than in July.

Number of unique users attacked by Trojan-Banker.AndroidOS.Asacub in Q2 and Q3 2017

New capabilities of mobile banking Trojans

Q3 2017 saw two significant events in the world of mobile banking Trojans.

Firstly, the family of mobile banking Trojans Svpeng has acquired the new modification capable of granting all the necessary rights to itself and stealing data from other applications. To do this, it just needs to persuade the user to allow the Trojan to utilize special functions designed for people with disabilities. As a result, the Trojan can intercept text that a user is entering, steal text messages and even prevent itself from being removed.

Interestingly, in August we discovered yet another modification of Svpeng that uses special features. Only, this time the Trojan was not banking related – instead of stealing data, it encrypts all the files on a device and demands a ransom in bitcoins. window containing ransom demand

Secondly, the FakeToken family of mobile banking Trojans has expanded the list of apps it attacks. If previously representatives of this family mostly overlaid banking and some Google apps (e.g. Google Play Store) with a phishing window, it is now also overlaying apps used to book taxis, air tickets and hotels. The aim of the Trojan is to harvest data from bank cards.

The growth of WAP billing subscriptions

In the third quarter of 2017, we continued to monitor the increased activity of Trojans designed to steal users’ money via subscriptions. To recap, these are Trojans capable of visiting sites that allow users to pay for services by deducting money from their mobile phone accounts. These Trojans can usually click buttons on such sites using special JS files, and thus make payments without the user’s knowledge.

Our Top 20 most popular Trojan programs in Q3 2017 included three malware samples that attack WAP subscriptions. They are Trojan-Dropper.AndroidOS.Agent.hb and Trojan.AndroidOS.Loapi.b in fourth and fifth, and Trojan-Clicker.AndroidOS.Ubsod.b in seventh place.

Mobile threat statistics

In the third quarter of 2017, Kaspersky Lab detected 1,598,196 malicious installation packages, which is 1.2 times more than in the previous quarter.

Number of detected malicious installation packages (Q4 2016 – Q3 2017)

Distribution of mobile malware by type

Distribution of new mobile malware by type (Q2 and Q3 2017)

RiskTool (53.44%) demonstrated the highest growth in Q3 2017, with its share increasing by 12.93 percentage points (p.p.). The majority of all installation packages discovered belonged to the RiskTool.AndroidOS.Skymobi family.

Trojan-Dropper malware (10.97%) came second in terms of growth rate: its contribution increased by 6.29 p.p. Most of the installation packages are detected as Trojan-Dropper.AndroidOS.Agent.hb.

The share of Trojan-Ransom programs, which was first in terms of the growth rate in the first quarter of 2017, continued to fall and accounted for 6.69% in Q3, which is 8.4 p.p. less than the previous quarter. The percentage of Trojan-SMS malware also fell considerably to 2.62% – almost 4 p.p. less than in Q2.

In Q3, Trojan-Clicker malware broke into this rating after its contribution increased from 0.29% to 1.41% in the space of three months.

TOP 20 mobile malware programs

Please note that this rating of malicious programs does not include potentially dangerous or unwanted programs such as RiskTool or adware.

Verdict % of attacked users* 1 DangerousObject.Multi.Generic 67.14 2 Trojan.AndroidOS.Boogr.gsh 7.52 3 4.56 4 Trojan-Dropper.AndroidOS.Agent.hb 2.96 5 Trojan.AndroidOS.Loapi.b 2.91 6 Trojan-Dropper.AndroidOS.Hqwar.i 2.59 7 Trojan-Clicker.AndroidOS.Ubsod.b 2.20 8 Backdoor.AndroidOS.Ztorg.c 2.09 9 2.05 10 Trojan.AndroidOS.Sivu.c 1.98 11 Trojan.AndroidOS.Hiddapp.u 1.87 12 Backdoor.AndroidOS.Ztorg.a 1.68 13 Trojan.AndroidOS.Agent.ou 1.63 14 Trojan.AndroidOS.Triada.dl 1.57 15 Trojan-Ransom.AndroidOS.Zebt.a 1.57 16 Trojan-Dropper.AndroidOS.Hqwar.gen 1.53 17 1.48 18 1.47 19 1.41 20 Trojan.AndroidOS.Agent.eb 1.29

* Percentage of unique users attacked by the malware in question, relative to all users of Kaspersky Lab’s mobile security product that were attacked.

First place was occupied by DangerousObject.Multi.Generic (67.14%), the verdict used for malicious programs detected using cloud technologies. This is basically how the very latest malware is detected.

As in the previous quarter, Trojan.AndroidOS.Boogr.gsh (7.52%) came second. This verdict is issued for files recognized as malicious by our system based on machine learning. (4.56%) was third. The main purpose of this Trojan is to open and click advertising links received from the C&C. The Trojan requests administrator rights to prevent its removal.

Trojan-Dropper.AndroidOS.Agent.hb (2.96%) climbed from sixth in Q2 to fourth this quarter. This Trojan decrypts and runs another Trojan – a representative of the Loaipi family. One of them –Trojan.AndroidOS.Loapi.b – came fifth in this quarter’s Top 20. This is a complex modular Trojan whose main malicious component needs to be downloaded from the cybercriminals’ server. We can assume that Trojan.AndroidOS.Loapi.b is designed to steal money via paid subscriptions.

Trojan-Dropper.AndroidOS.Hqwar.i (3.59%), the verdict used for Trojans protected by a certain packer/obfuscator, fell from fourth to sixth. In most cases, this name indicates representatives of the FakeToken and Svpeng mobile banking families.

In seventh was Trojan-Clicker.AndroidOS.Ubsod.b, a small basic Trojan that receives links from a C&C and opens them. We wrote about this family in more detail in our review of Trojans that steal money using WAP subscriptions.

Trojan Backdoor.AndroidOS.Ztorg.c came eighth. This is one of the most active advertising Trojans that uses superuser rights. In the third quarter of 2017, our Top 20 included eight Trojans that try to obtain or use root rights and which make use of advertising as their main means of monetization. Their goal is to deliver ads to the user more aggressively, applying (among other methods) hidden installation of new advertising programs. At the same time, superuser privileges help them ‘hide’ in the system folder, making it very difficult to remove them. It’s worth noting that the quantity of this type of malware in the Top 20 has been decreasing (in Q1 2017, there were 14 of these Trojans in the rating, while in Q2 the number was 11). (2.05%), which steals money from users making calls to premium numbers, rose from fifteenth to ninth. Due to its use of administrator rights, it resists attempts to remove it from an infected device.

Occupying fifteenth this quarter was Trojan-Ransom.AndroidOS.Zebt.a, the first ransom Trojan in this Top 20 rating in 2017. This is a fairly simple Trojan whose main goal is to block the device with its window and demand a ransom. Zebt.a tends to attack users in Europe and Mexico. (1.48%) fell to sixteenth after occupying second and third in the previous two quarters. This piece of malware imitates various popular games or programs. Interestingly, once run, it downloads and installs the application it imitated. In this case, the Trojan requests administrator rights to withstand removal. The main purpose of is the aggressive display of adverts. Its main ‘audience’ is in Russia.

The geography of mobile threats

The geography of attempted mobile malware infections in Q3 2017 (percentage of all users attacked)

Top 10 countries attacked by mobile malware (ranked by percentage of users attacked):

Country* % of attacked users** 1 Iran 35.12 2 Bangladesh 28.30 3 China 27.38 4 Côte d’Ivoire 26.22 5 Algeria 24.78 6 Nigeria 23.76 7 Indonesia 22.29 8 India 21.91 9 Nepal 20.78 10 Kenya 20.43

* We eliminated countries from this rating where the number of users of Kaspersky Lab’s mobile security product is relatively low (under 10,000). 
** Percentage of unique users attacked in each country relative to all users of Kaspersky Lab’s mobile security product in the country.

For the third quarter in a row Iran was the country with the highest percentage of users attacked by mobile malware – 35.12%. Bangladesh came second, with 28.3% of users there encountering a mobile threat at least once during Q3. China (27.38%) followed in third.

Russia (8.68%) came 35th this quarter (vs 26th place in Q2), France (4.9%) was 59th, the US (3.8%) 67th, Italy (5.3%) 56th, Germany (2.9%) 79th, and the UK (3.4%) 72nd.

The safest countries were Georgia (2.2%), Denmark (1.9%), and Japan (0.8%).

Mobile banking Trojans

Over the reporting period we detected 19,748 installation packages for mobile banking Trojans, which is 1.4 times less than in Q2 2017.

Number of installation packages for mobile banking Trojans detected by Kaspersky Lab solutions (Q4 2016 – Q3 2017) became the most popular mobile banking Trojan in Q3, replacing the long-term leader Trojan-Banker.AndroidOS.Svpeng.q. These mobile banking Trojans use phishing windows to steal credit card data and logins and passwords for online banking accounts. In addition, they steal money via SMS services, including mobile banking.

Geography of mobile banking threats in Q3 2017 (percentage of all users attacked)

Top 10 countries attacked by mobile banker Trojans (ranked by percentage of users attacked):

Country* % of attacked users** 1 Russia 1.20 2 Uzbekistan 0.40 3 Kazakhstan 0.36 4 Tajikistan 0.35 5 Turkey 0.34 6 Moldova 0.31 7 Ukraine 0.29 8 Kyrgyzstan 0.27 9 Belarus 0.26 10 Latvia 0.23

* We eliminated countries from this rating where the number of users of Kaspersky Lab’s mobile security product is relatively low (under 10,000).
** Percentage of unique users in each country attacked by mobile banker Trojans, relative to all users of Kaspersky Lab’s mobile security product in the country.

In Q3 2017, the Top 10 countries attacked by mobile banker Trojans saw little change: Russia (1.2%) topped the ranking again. In second and third places were Uzbekistan (0.4%) and Kazakhstan (0.36%), which came fifth and tenth respectively in the previous quarter. In these countries the Faketoken.z, Tiny.b and Svpeng.y families were the most widespread threats.

Of particular interest is the fact that Australia, a long-term resident at the top end of this rating, didn’t make it into our Top 10 this quarter. This was due to a decrease in activity by the Trojan-Banker.AndroidOS.Acecard and Trojan-Banker.AndroidOS.Marcher mobile banking families.

Mobile ransomware

In Q3 2017, we detected 108,073 mobile Trojan-Ransomware installation packages, which is almost half as much as in the previous quarter.

Number of mobile Trojan-Ransomware installation packages detected by Kaspersky Lab (Q3 2016 – Q3 2017)

In our report for Q2, we wrote that in the first half of 2017, we had discovered more mobile ransomware installation packages than in any other period. The reason was the Trojan-Ransom.AndroidOS.Congur family. However, in the third quarter of this year we observed a decline in this family’s activity.

Trojan-Ransom.AndroidOS.Zebt.a became the most popular mobile Trojan-Ransomware in Q3, accounting for more than a third of users attacked by mobile ransomware. Second came Trojan-Ransom.AndroidOS.Svpeng.ab. Meanwhile, Trojan-Ransom.AndroidOS.Fusob.h, which topped the rating for several quarters in a row, was only third in Q3 2017.

Geography of mobile Trojan-Ransomware in Q3 2017 (percentage of all users attacked)

Top 10 countries attacked by mobile Trojan-Ransomware (ranked by percentage of users attacked):

1 US 1.03% 2 Mexico 0.91% 3 Belgium 0.85% 4 Kazakhstan 0.79% 5 Romania 0.70% 6 Italy 0.50% 7 China 0.49% 8 Poland 0.49% 9 Austria 0.45% 10 Spain 0.33%

* We eliminated countries from this ranking where the number of users of Kaspersky Lab’s mobile security product is lower than 10,000.
** Percentage of unique users in each country attacked by mobile Trojan-Ransomware, relative to all users of Kaspersky Lab’s mobile security product in the country.

The US (1.03%) again topped the rating of countries attacked most by mobile Trojan-Ransomware; the most widespread family in the country was Trojan-Ransom.AndroidOS.Svpeng. These Trojans appeared in 2014 as a modification of the Trojan-Banker.AndroidOS.Svpeng mobile banking family. They demand a ransom of about $500 from victims to unblock their devices.

In Mexico (0.91%), which came second in Q3 2017, most mobile ransomware attacks involved Trojan-Ransom.AndroidOS.Zebt.a. Belgium (0.85%) came third, with Zebt.a the main threat to users there too.

Vulnerable apps exploited by cybercriminals

Q3 2017 saw continued growth in the number of attacks launched against users involving malicious Microsoft Office documents. We noted the emergence of a large number of combined documents containing an exploit as well as a phishing message – in case the embedded exploit fails.

Although two new Microsoft Office vulnerabilities, CVE-2017-8570 and CVE-2017-8759, have emerged, cybercriminals have continued to exploit CVE-2017-0199, a logical vulnerability in processing HTA objects that was discovered in March 2017. Kaspersky Lab statistics show that attacks against 65% users in Q3 exploited CVE-2017-0199, and less than 1% exploited CVE-2017-8570 or CVE-2017-8759. The overall share of exploits for Microsoft Office was 27.8%.

There were no large network attacks (such as WannaCry or ExPetr) launched in Q3 using vulnerabilities patched by the MS17-010 update. However, according to KSN data, there was major growth throughout the quarter in the number of attempted exploitations of these vulnerabilities that were blocked by our Intrusion Detection System component. Unsurprisingly, the most popular exploits have been EternalBlue and its modifications, which use an SMB protocol vulnerability; however, KL statistics show that EternalRomance, EternalChampion and an exploit for the CVE-2017-7269 vulnerability in IIS web servers have also been actively used by cybercriminals. EternalBlue, however, accounts for millions of blocked attempted attacks per month, while the numbers for other exploits are much lower.

Distribution of exploits used in attacks by type of application attacked, Q3 2017

The distribution of exploits by the type of attacked application this quarter was practically the same as in Q2. First place is still occupied by exploits targeting browsers and browser components with a share of 35.0% (a decline of 4 p.p. compared to Q2.) The proportion of exploits targeting Android vulnerabilities (22.7%) was almost identical to that in Q2, placing this type of attacked application once again in third behind Office vulnerabilities.

Online threats (Web-based attacks)

These statistics are based on detection verdicts returned by the web antivirus module that protects users at the moment when malicious objects are downloaded from a malicious/infected web page. Malicious sites are specifically created by cybercriminals; infected web resources include those whose content is created by users (e.g. forums), as well as legitimate resources.

Online threats in the banking sector

These statistics are based on detection verdicts of Kaspersky Lab products, received from users of Kaspersky Lab products who have consented to provide their statistical data. Beginning from the first quarter of 2017 these statistics include malicious programs for ATMs and POS terminals, but do not include mobile threats.

In Q3 2017, Kaspersky Lab solutions blocked attempts to launch one or more malicious programs capable of stealing money via online banking on 204,388 computers.

Number of users attacked by financial malware, Q3 2017

Geography of attacks

To evaluate and compare the risk of being infected by banking Trojans and ATM and POS-malware worldwide, we calculate the percentage of Kaspersky Lab product users in the country who encountered this type of threat during the reporting period, relative to all users of our products in that country.

Geography of banking malware attacks in Q3 2017 (percentage of all users attacked)

TOP 10 countries attacked by mobile banker Trojans (ranked by percentage of users attacked)

Country* % of users attacked** 1 Togo 2.30 2 China 1.91 3 Taiwan 1.65 4 Indonesia 1.58 5 South Korea 1.56 6 Germany 1.53 7 United Arab Emirates 1.52 8 Lebanon 1.48 9 Libya 1.43 10 Jordan 1.33

These statistics are based on detection verdicts returned by the antivirus module, received from users of Kaspersky Lab products who have consented to provide their statistical data.
* We excluded those countries in which the number of Kaspersky Lab product users is relatively small (under 10,000).
** Unique users whose computers have been targeted by banking Trojan malware attacks as a percentage of all unique users of Kaspersky Lab products in the country.

TOP 10 banking malware families

The table below shows the Top 10 malware families used in Q3 2017 to attack online banking users (in terms of percentage of users attacked):

Name* % of attacked users** 1 Trojan-Spy.Win32.Zbot 27.9 2 Trojan.Win32.Nymaim 20.4 3 Trojan.Win32.Neurevt 10.0 4 Trickster 9.5 5 SpyEye 7.5 6 Caphaw 6.3 7 Trojan-Banker.Win32.Gozi 2.0 8 Shiz 1.8 9 ZAccess 1.6 10 NeutrinoPOS 1.6

* The detection verdicts of Kaspersky Lab products, received from users of Kaspersky Lab products who have consented to provide their statistical data.
** Unique users whose computers have been targeted by the malware in question as a percentage of all users attacked by financial malware.

The malware families Dridex and Tinba lost their places in this quarter’s Top 10. One of their former positions was occupied by the Trickster bot (accounting for 9.5% of attacked users), also known as TrickBot, a descendant of the now defunct Dyre banker. There was a small change in the leading three malicious families. First and second places are still occupied by Trojan-Spy.Win32.Zbot (27.9%) and Trojan.Win32.Nymaim (20.4%) respectively, while third place is now occupied by Trojan.Win32.Neurevt (10%) whose share grew by nearly 4 p.p.

Cryptoware programs Q3 highlights Crysis rises from the dead

In our Q2 report we wrote that the cybercriminals behind the Crysis ransomware cryptor halted distribution of the malware and published the secret keys needed to decrypt files. This took place in May 2017, and all propagation of the ransomware was stopped completely at that time.

However, nearly three months later, in mid-August, we discovered that this Trojan had come back from the dead and had set out on a new campaign of active propagation. The email addresses used by the blackmailers were different from those used in earlier samples of Crysis. A detailed analysis revealed that the new samples of the Trojan were completely identical to the old ones apart from just one thing – the public master keys were new. Everything else was the same, including the compilation timestamp in the PE header and, more interestingly, the labels that the Trojan leaves in the service area at the end of each encrypted file. Closer scrutiny of the samples suggests that the new distributors of the malware didn’t have the source code, so they just took its old body and used a HEX editor to change the key and the contact email.

The above suggests that this piece of ‘zombie’ malware is being spread by a different group of malicious actors rather than its original developer who disclosed all the private keys in May.

Surge in Cryrar attacks

The Cryrar cryptor (aka ACCDFISA) is a veteran among the ransomware Trojans that are currently being spread. It emerged way back in 2012 and has been active ever since. The cryptor is written in PureBasic and uses a legitimate executable RAR archiver file to place the victim’s files in password-encrypted RAR-sfx archives.

In the first week of September 2017 we recorded a dramatic rise in the number of attempted infections with Cryrar – a surge never seen before or since. The malicious actors used the following approach: they crack the password to RDP by brute force, get authentication on the victim’s system using the remote access protocol and manually launch the Trojan’s installation file. The latter, in turn, installs the cryptor’s body and the components it requires (including the renamed RAR.EXE file), and then automatically launches the cryptor.

According to KSN data, this wave of attacks primarily targeted Vietnam, China, the Philippines and Brazil.

Master key to original versions of Petya/Mischa/GoldenEye published

In July 2017, the authors of the Petya Trojan published their master key, which can be used to decrypt the Salsa keys required to decrypt MFT and unblock access to systems affected by Petya/Mischa or GoldenEye.

This happened shortly after the ExPetr epidemic which used part of the GoldenEye code. This suggests that the authors of Petya/Mischa/GoldenEye did so in an attempt to distance themselves from the ExPetr attack and the outcry that it caused.

Unfortunately, this master key won’t help those affected by ExPetr, as its creators didn’t include the option of restoring a Salsa key to decrypt MFT.

The number of new modifications

In Q3 2017, we identified five new ransomware families in this classification. It’s worth noting here that this number doesn’t include all the Trojans that weren’t assigned their own ‘personal’ verdict. Each quarter, dozens of these malicious programs emerge, though they either have so few distinctive characteristics or occur so rarely that they and the hundreds of others like them remain nameless, and are detected with generic verdicts.

Number of newly created cryptor modifications, Q3 2016 – Q3 2017

The number of new cryptor modifications continues to decline compared to previous quarters. This could be a temporary trend, or could indicate that cybercriminals are gradually losing their interest in cryptors as a means of making money, and are switching over to other types of malware.

The number of users attacked by ransomware

July was the month with the lowest ransomware activity. From July to September, the number of ransomware attacks rose, though it remained lower than May and June when two massive epidemics (WannaCry and ExPetr) struck.

Number of unique users attacked by Trojan-Ransom cryptor malware (Q3 2017)

The geography of attacks

Top 10 countries attacked by cryptors Country* % of users attacked by cryptors** 1 Myanmar 0.95% 2 Vietnam 0.92% 3 Indonesia 0.69% 4 Germany 0.62% 5 China 0.58% 6 Russia 0.51% 7 Philippines 0.50% 8 Venezuela 0.50% 9 Cambodia 0.50% 10 Austria 0.49%

* We excluded those countries where the number of Kaspersky Lab product users is relatively small (under 50,000)
** Unique users whose computers have been targeted by ransomware as a percentage of all unique users of Kaspersky Lab products in the country.

Most of the countries in this Top 10 are from Asia, including Myanmar (0.95%), a newcomer to the Top 10 that swept into first place in Q3. Vietnam (0.92%) came second, moving up two places from Q2, while China (0.58%) rose one place to fifth.

Brazil, Italy and Japan were the leaders in Q2, but in Q3 they failed to make it into the Top 10. Europe is represented by Germany (0.62%) and Austria (0.49%).

Russia, in tenth the previous quarter, ended Q3 in sixth place.

Top 10 most widespread cryptor families Name Verdict* % of attacked users** 1 WannaCry Trojan-Ransom.Win32.Wanna 16.78% 2 Crypton Trojan-Ransom.Win32.Cryptoff 14.41% 3 Purgen/GlobeImposter Trojan-Ransom.Win32.Purgen 6.90% 4 Locky Trojan-Ransom.Win32.Locky 6.78% 5 Cerber Trojan-Ransom.Win32.Zerber 4.30% 6 Cryrar/ACCDFISA Trojan-Ransom.Win32.Cryrar 3.99% 7 Shade Trojan-Ransom.Win32.Shade 2.69% 8 Spora Trojan-Ransom.Win32.Spora 1.87% 9 (generic verdict) Trojan-Ransom.Win32.Gen 1.77% 10 (generic verdict) Trojan-Ransom.Win32.CryFile 1.27%

* These statistics are based on detection verdicts received from users of Kaspersky Lab products who have consented to provide their statistical data.
** Unique users whose computers have been targeted by a specific Trojan-Ransom family as a percentage of all users of Kaspersky Lab products attacked by Trojan-Ransom malware.

Wannacry (16.78%) tops the rating for Q3, and the odds are that it’s set to remain there: the worm has been propagating uncontrollably, and there are still huge numbers of computers across the globe with the unpatched vulnerability that Wannacry exploits.

Crypton (14.41%) came second. This cryptor emerged in spring 2016 and has undergone many modifications since. It has also been given multiple names: CryptON, JuicyLemon, PizzaCrypts, Nemesis, x3m, Cry9, Cry128, Cry36.

The cryptor Purgen (6.90%) rounds off the top three after rising from ninth. The rest of the rating is populated by ‘old timers’ – the Trojans Locky, Cerber, Cryrar, Shade, and Spora.

The Jaff cryptor appeared in the spring of 2017, going straight into fourth place in the Q2 rating, and then stopped spreading just as suddenly.

Top 10 countries where online resources are seeded with malware

The following statistics are based on the physical location of the online resources used in attacks and blocked by our antivirus components (web pages containing redirects to exploits, sites containing exploits and other malware, botnet command centers, etc.). Any unique host could be the source of one or more web attacks. In order to determine the geographical source of web-based attacks, domain names are matched against their actual domain IP addresses, and then the geographical location of a specific IP address (GEOIP) is established.

In the third quarter of 2017, Kaspersky Lab solutions blocked 277,646,376 attacks launched from web resources located in 185 countries around the world. 72,012,219 unique URLs were recognized as malicious by web antivirus components.

Distribution of web attack sources by country, Q3 2017

In Q3 2017, the US (3.86%) was home to most sources of web attacks. The Netherlands (25.22%) remained in second place, while Germany moved up from fifth to third. Finland and Singapore dropped out of the top five and were replaced by Ireland (1.36%) and Ukraine (1.36%).

Countries where users faced the greatest risk of online infection

In order to assess the risk of online infection faced by users in different countries, we calculated the percentage of Kaspersky Lab users in each country who encountered detection verdicts on their machines during the quarter. The resulting data provides an indication of the aggressiveness of the environment in which computers work in different countries.

This rating only includes attacks by malicious programs that fall under the Malware class. The rating does not include web antivirus module detections of potentially dangerous or unwanted programs such as RiskTool or adware.

Country* % of users attacked** 1 Belarus 27.35 2 Algeria 24.23 3 Russia 23.91 4 Armenia 23.74 5 Moldova 23.61 6 Greece 21.48 7 Azerbaijan 21.14 8 Kyrgyzstan 20.83 9 Uzbekistan 20.24 10 Albania 20.10 11 Ukraine 19.82 12 Kazakhstan 19.55 13 France 18.94 14 Venezuela 18.68 15 Brazil 18.01 16 Portugal 17.93 17 Vietnam 17.81 18 Tajikistan 17.63 19 Georgia 17.50 20 India 17.43

These statistics are based on detection verdicts returned by the web antivirus module, received from users of Kaspersky Lab products who have consented to provide their statistical data.
* These calculations excluded countries where the number of Kaspersky Lab users is relatively small (under 10,000 users).
** Unique users whose computers have been targeted by Malware-class attacks as a percentage of all unique users of Kaspersky Lab products in the country.

On average, 16.61% of computers connected to the Internet globally were subjected to at least one Malware-class web attack during the quarter.

Geography of malicious web attacks in Q3 2017 (ranked by percentage of users attacked)

The countries with the safest online surfing environments included Iran (9.06%), Singapore (8.94%), Puerto Rico (6.67%), Niger (5.14%) and Cuba (4.44%).

Local threats

Local infection statistics for user computers are a very important indicator: they reflect threats that have penetrated computer systems by infecting files or removable media, or initially got on the computer in an encrypted format (for example, programs integrated in complex installers, encrypted files, etc.).

Data in this section is based on analyzing statistics produced by antivirus scans of files on the hard drive at the moment they were created or accessed, and the results of scanning removable storage media.

In Q3 2017, Kaspersky Lab’s file antivirus detected 198,228,428 unique malicious and potentially unwanted objects.

Countries where users faced the highest risk of local infection

For each country, we calculated the percentage of Kaspersky Lab product users on whose computers the file antivirus was triggered during the quarter. These statistics reflect the level of personal computer infection in different countries.

The rating of malicious programs only includes Malware-class attacks. The rating does not include web antivirus module detections of potentially dangerous or unwanted programs such as RiskTool or adware.

Country* % of users attacked** 1 Yemen 56.89 2 Vietnam 54.32 3 Afghanistan 53.25 4 Uzbekistan 53.02 5 Laos 52.72 6 Tajikistan 49.72 7 Ethiopia 48.90 8 Syria 47.71 9 Myanmar 46.82 10 Cambodia 46.69 11 Iraq 45.79 12 Turkmenistan 45.47 13 Libya 45.00 14 Bangladesh 44.54 15 China 44.40 16 Sudan 44.27 17 Mongolia 44.18 18 Mozambique 43.84 19 Rwanda 43.22 20      Belarus 42.53

These statistics are based on detection verdicts returned by on-access and on-demand antivirus modules, received from users of Kaspersky Lab products who have consented to provide their statistical data. The data include detections of malicious programs located on users’ computers or on removable media connected to the computers, such as flash drives, camera and phone memory cards, or external hard drives.
* These calculations exclude countries where the number of Kaspersky Lab users is relatively small (under 10,000 users).
** The percentage of unique users in the country with computers that blocked Malware-class local threats as a percentage of all unique users of Kaspersky Lab products.

This Top 20 of countries has not changed much since Q2, with the exception of China (44.40%), Syria (47.71%) and Libya (45.00%) all making an appearance. The proportion of users attacked in Russia amounted to 29.09%.

On average, 23.39% of computers globally faced at least one Malware-class local threat during the third quarter.

Geography of local malware attacks in Q3 2017 (ranked by percentage of users attacked)

The safest countries in terms of local infection risks included Estonia (15.86%), Singapore (11.97%), New Zealand (9.24%), Czechia (7.89%), Ireland (6.86%) and Japan (5.79%).

All the statistics used in this report were obtained using Kaspersky Security Network (KSN), a distributed antivirus network that works with various anti-malware protection components. The data was collected from KSN users who agreed to provide it. Millions of Kaspersky Lab product users from 213 countries and territories worldwide participate in this global exchange of information about malicious activity.

IT threat evolution Q3 2017

Fri, 11/10/2017 - 05:09

Targeted attacks and malware campaigns [Re-]enter the dragon

In July, we reported on the recent activities of a targeted attack group called ‘Spring Dragon’ (also known as LotusBlossom), whose activities data back to 2012. Spring Dragon makes extensive use of spear-phishing and watering-hole attacks. The group’s targets include high-profile government agencies, political parties, educational institutions and telecommunication around the South China Sea – including Taiwan, Indonesia, Vietnam, the Philippines, Hong Kong, Malaysia and Thailand.

Most of the malicious tools implemented by Spring Dragon over the years are backdoors designed to steal data, execute additional malware components and run system commands on victim’s computers. These give the attackers the ability to undertake a variety of different malicious activities on their victims’ computers. The group maintains a large C2 infrastructure, comprising more than 200 unique IP addresses and C2 domains.

The large number of samples that we have collected have customized configuration data, different sets of C2 addresses with new hardcoded campaign IDs, as well as customized configuration data for creating a service for malware on a victim’s system – all of which makes detection more difficult.

We think it is likely that Spring Dragon, like many other targeted attack campaigns, is likely to re-surface in this region, so it is important for organisations to make effective use of good detection mechanisms such as YARA rules and IDS signatures.

You can read our report on Spring Dragon here.


One of the most striking aspects of the ExPetr attacks earlier this year was its primary attack vector: the attackers specifically targeted a company supplying accounting software to Ukrainian companies. Most of the victims of this wiper were located in Ukraine. However, it recently became clear that the attack has had a significant impact on some companies that operate worldwide. Among them are Maersk, the world’s largest container ship and supply vessel company. The company indicated in its earnings report that it expected losses of between $200 and $300 as a result of ‘significant business interruption’ caused by the ExPetr attack. Another was FedEx, which revealed that the operations of its TNT Express unit in Europe were ‘significantly affected’ by the attack, costing the company around $300 in lost earnings.

In recent months, we have seen further cases of attackers compromising software supply chain providers and using this as a stepping-stone into their chosen targets.

In July, we discovered suspicious DNS requests on the network of a customer working in the financial services industry: we found the requests on systems used to process transactions. The source of the DNS queries was a package for popular server management software developed by NetSarang. Customers of NetSarang, which has headquarters in South Korea and the United States, include companies working in financial services, energy, retail, technology and media. The attackers had modified one of the updates to include a backdoor.

NetSarang quickly removed the compromised update, but not before it had been activated at least once (we were able to confirm an activation on a computer in Hong Kong).

The attackers hide their malicious intent in several layers of encrypted code. The tiered architecture means that the business logic of the backdoor is not activated until a special packet has been received from the first tier C2 (Command and Control) server. Until then, it transfers basic information every eight hours: this includes computer, domain and user names. The payload is only activated through a crafted ‘dns.txt’ record for a specific domain. This allows the attackers to glean system information and send a decryption key to unlock the next stage of the attack, activating the backdoor itself.

This backdoor, called ShadowPad, is a modular platform that lets the attackers download and execute arbitrary code, create processes and maintain a virtual file system in the registry, all of which are encrypted and stored in locations unique to each victim.

You can read more about ShadowPad here.

Another supply-chain attack occurred in September, when attackers compromised an update to the Windows clean-up utility CCleaner, published by Avast. Researchers at Cisco Systems Talos Group discovered that attackers had modified the installer for CCleaner 5.3 to drop their malware on the computers of anyone who downloaded the utility. The malware, which was signed with a valid certificate, was active for a month and infected around 700,000 computers. The attackers used a two-stage infection process. The first delivered a profile of the victim to the attackers C2 servers, while the second was reserved for specific targets. You can find details of the analysis here.

It is sometimes tempting for companies to imagine that no one would want to target them – perhaps because they are not a large company, or because they do not believe that they have anything of significance to an attacker. However, even quite apart from their intellectual property, or personal information belonging to customers, they can be valuable as a stepping-stone into another organisation.

The bear facts

In August, we provided an update on an interesting APT that we call ‘WhiteBear’, related to the Turla group. Like Turla, WhiteBear uses compromised web sites and hijacked satellite connections for its C2 infrastructure. The project also overlaps with other Turla campaigns such as ‘Skipper Turla’ (or ‘WhiteAtlas’) and ‘Kopiluwak’ (both of which we detailed for subscribers to Kaspersky APT intelligence reports). In addition, we have found WhiteBear components on a subset of systems that were previously targeted by WhiteAtlas, with the same file-paths and identical filenames. Nevertheless, we have been unable to firmly tie the delivery of WhiteBear to any specific WhiteAtlas components, and we believe that WhiteBear is the product of a separate development effort and has a distinct focus.

For much of 2016, WhiteBear activity was narrowly focused on embassies and consulates around the world – all related to diplomatic and foreign affairs organisations. This shifted in mid-2017 to include defence-related organizations.

Although we’re not sure of the delivery vector for WhiteBear components, we strongly suspect that the group sends spear-phishing e-mails to its targets containing malicious PDF files.

The encryption implemented in the main module, the WhiteBear orchestrator, is particularly interesting. The attackers encrypt/decrypt, and pack/decompress the resource section with RSA+3DES+BZIP2. This implementation is unique and includes the format of the private key as stored in the resource section. 3DES is also present in Sofacy and Duqu 2.0 components, but they are missing in this Microsoft-centric RSA encryption technique. The private key format used in this schema and the RSA crypto combination with 3DES is (currently) unique to this group.

Most WhiteBear samples are signed with a valid code-signing certificate issued for ‘Solid Loop Ltd’, a once-registered British organization. This is probably a front organization or a defunct organization; and the attackers have assumed its identity to abuse the name and trust, in order to create deceptive digital certificates.

You can find full technical details of WhiteBear here.

(Un)documented Word feature abused by hackers

If a targeted attack is to be successful, the attackers must first gather intelligence on their prospective victims. In particular, they need details about the operating system and key applications, so that they can deliver the appropriate exploit.

During an investigation of a targeted attack, we found some spear-phishing e-mails with interesting Word documents attached to them. At first sight, they seemed unremarkable: they contained no macros, exploits or other active content.

However, on closer inspection, we found that they contained several links to PHP scripts located on third-party web resources. When we attempted to open these files in Microsoft Word, we found that the application addressed one of the links and, as a result, provided the attackers with information about software installed on the target computer. The documents were in OLE 2 (Object Linking and Embedding) format. OLE allows authors to embed objects and link to multiple objects or resources in a single Word document. For example, an author can created a field in a document that points to a graphic file, rather than simply embedding the graphic file.

We found a field in the document called ‘INCLUDEPICTURE’. The link to the image in this field should be in ASCII, but in this case, it was in Unicode. Microsoft documentation provides virtually no information about this field. However, the attackers manipulated the Unicode framework to trigger a GET request to malicious and obfuscated URLs contained in the underlying code of the Word document. These links then point to PHP scripts located on third-party web sites, enabling the attackers to gather information about the software installed on the computer.

This feature is not only present in Word for Windows, but also in Microsoft Office for iOS and in Microsoft Office for Android.

You can read further details about our investigation here.

Information security incidents and how to respond to them

Our growing dependence on technology, connectivity and data means that businesses present a bigger attack surface than ever. Targeted attackers have become more adept at exploiting their victims’ vulnerabilities to penetrate corporate defences while ‘flying under the radar’. Unfortunately, corporate information security services are often unprepared. Their employees underestimate the speed, secrecy and efficiency of modern cyber-attacks and businesses often fail to recognize how ineffective the old approaches to security are. Even where companies supplement traditional prevention tools such as anti-malware products, IDS/IPS and security scanners with detection solutions such as SIEM and anti-APT, they may not be used to their full potential.

You can’t manage what you can’t measure. One of the key factors in responding effectively to a targeted attack is to understand the nature of the incident.

In August, our incident response team used the example of a bank attack to present the key stages of a targeted attack (known as the kill chain) and the steps required for an effective incident response process. You can read the report here, but the following is a summary of the key elements.

The basic principles of a successful targeted attack include thorough preparation and a step-by-step strategy. The stages of the kill chain are:

  1. RECONNAISSANCE (learning about the target)
  2. WEOPANISATION (choosing the method of attack)
  3. DELIVERY (deciding on the attack vector)
  4. EXPLOITATION (exploiting a vulnerability to gain an initial foothold)
  5. INSTALLATION (installing the malware)
  6. COMMAND-AND-CONTROL (connecting to the attackers’ server for further instructions)
  7. ACTIONS ON OBJECTIVE (achieving the attackers’ goals)

The basic principles behind the work of information security staff are the same as the attackers – careful preparation and a step-by-step strategy. The objectives, of course, are fundamentally different: to prevent incidents and, if one occurs, to restore the initial state of the system as soon as possible.

There are two main stages involved in responding to a specific incident: investigation and system restoration. The investigation must determine

  • The initial attack vector
  • The malware, exploits and other tools use by the attackers
  • The target of the attack (affected networks, systems and data)
  • The extent of the damage (including reputational damage) to the organisation
  • The stage of the attack (whether or not it was completed and the attackers’ goals were achieved)
  • Timeframes (when the attack started and ended, when it was detected and the response time of the information security service)

Once the investigation has been completed, it is necessary to use the information learned to create a system recovery plan or, if one exists, to assess how it can be improved.

The overall strategy includes the following steps.

  1. PREPARATION (develop the tools, policies and processes needed to defend the organisation)
  2. IDENTIFICATION (decide if an incident has occurred by identifying pre-defined triggers)
  3. CONTAINMENT (limit the scope of the incident and maintain business continuity)
  4. ERADICATION (restore the system to its pre-incident state)
  5. RECOVERY (re-connect the affected systems to the wider network)
  6. LESSONS LEARNED (how well did the information security team deal with the incident and what changes need to be made to the strategy)

In the event of the information security team having to respond to multiple incidents simultaneously, it’s important to correctly set priorities and focus on the main threats. The key factors involved in determining the severity of an incident include:

  • The network segment where the compromised computer is located
  • The value of the data stored on that computer
  • The type and number of incidents that affect the same computer
  • The reliability of the IoCs (Indicators of Compromise) for this incident

The choice of computer, server or network segment to deal with first will depend on the specific nature of the organisation.

Malware stories The hidden advertising threat

As well as banking Trojans, ransomware and other threats that can clearly be defined as malware, people also face numerous borderline programs – including advertising bots and modules, and partnership programs – which are typically referred to as ‘potentially unwanted programs’. They are borderline because there is sometimes a fine line between classifying something as an outright Trojan or adware. One such program is Magala, a Trojan-Clicker.

Such programs imitate a user click on a particular web page, thus boosting advertisement click counts. Magala doesn’t actually affect the person whose computer it is installed on, other than consuming some of their computer’s resources. The victims are those who pay for the advertising – typically small business owners doing business with unscrupulous advertisers.

The first stage of the infection involves the Trojan checking which version of Internet Explorer is installed and locating it in the system. The Trojan doesn’t run if it’s version 8 or earlier. Otherwise, it initialises a virtual desktop, used to perform all subsequent activities. Then it runs a sequence of utility operations (typical for this type of malware): it sets up autorun, sends a report to a hardcoded URL, and installs the required adware. To interact with the content of an open page, Magala uses IHTMLDocument2, the standard Windows interface that makes it easy to use DOM tree. The Trojan uses it to load the MapsGalaxy Toolbar, installs this on the system and adds the site ‘hxxp://’ to the system registry, associating it with MapsGalaxy so that it becomes the browser’s home page.

The Trojan then contacts the remote server and requests a list of search queries for the click counts that it needs to boost. The server returns this list in plain text. Magala uses the list to send the requested search queries and clicks on each of the first 10 links in the search results, with an interval of 10 seconds between each click.

The average cost per click in a campaign of this sort is $0.07. So a botnet consisting of 1,000 infected computers clicking 10 web site addresses from each search result, performing 500 search requests with no overlaps in the search results, could earn the cybercriminals up to $350 from each infected computer. However, this is just an estimate as the costs can vary greatly in each situation.

Statistics from March to early June 2017 indicate that most Magala infections occur in the United States and Germany.

This class of program typically doesn’t present as much of a threat to consumers as, for example, banking Trojans or ransomware. However, two things make it tricky to deal with. First, such programs straddle the borderline between legitimate and malicious software and it’s vital to determine whether a specific program is part of a secure and legal advertising campaign or if it’s illegitimate software making use of similar functions. Second, the sheer quantity of such programs means that we need to use a fundamentally different approach to analysis.

You can read more about Magala here.

It started with a link

Cybercriminals are constantly on the lookout for ways of luring unsuspecting victims into doing things that compromise their security and capture personal data. In August, David Jacoby from Kaspersky Lab and Frans Rosen from Detectify teamed up to expose one such campaign that used Facebook Messenger to infect people.

It started with a link to a YouTube video. The cybercriminals behind the scam used social engineering to trick their victims into clicking on it: the message contained the recipient’s first name, plus the word ‘Video’ – for example ‘David Video’ – and then a link.

This link pointed to Google Drive, where the victim would see what looks like a playable movie, with a picture of them in the background and what seems to be a ‘Play’ button.

If the victim tried to play the video in the Chrome browser, they were redirected to what looked like a YouTube video and were prompted to install a Chrome extension –in fact, this was the malware. The malware waited for the victim to sign in to their Facebook account and stole their login credentials. It also captured information about their Facebook contacts and sent malicious links to their friends – so spreading the infection further.

Anyone using a different extension was nagged into updating their Adobe Flash Player instead – but the file they downloaded was adware, earning money for the cybercriminals through advertising.

This attack relied heavily on realistic social interactions, dynamic user content and legitimate domains as middle steps. The core infection point of the spreading mechanism was the installation of a Chrome Extension. It’s really important to be careful about allowing extensions to control your browser interactions and also to make sure that you know exactly what extensions you are running in your browser. In Chrome, you can type ‘chrome://extensions/’ into the address field of your browser to get a list of enabled extensions. On top of this, of course, be wary about clicking on links. If you’re in any doubt about whether it’s legitimate or not, contact the sender to check if it was really them who sent it.

Undermining your security

We have seen a substantial growth in crypto-currency miners this year. In 2013, our products blocked attempts to install miners on the computers of 205,000 people protected by Kaspersky Lab products. In 2014, this increased to 701,000. In the first eight months of 2017, this increased to 1.65 million.

Crypto-currency mining is not illegal. However, there are groups of people who trick unwitting people into installing mining software on their computers, or exploit software vulnerabilities to do so. The criminals obtain crypto-currency, while the computers of their victims slow down. We have recently detected several large botnets designed to profit from concealed crypto mining. We have also seen growing numbers of attempts to install miners on servers owned by organizations. When these attempts are successful, the business processes of the target organisations suffer because data processing speeds fall substantially.

The main method used to install miners is adware installers spread using social engineering. There are also more sophisticated propagation methods – one is using the EternalBlue exploit published in April 2017 by the Shadow Brokers group. In this case, the cybercriminals tend to target servers – these provide them with a more powerful asset.

We recently detected a network made up of an estimated 5,000 plus computers on which Minergate, a legal console miner, had been installed without the knowledge or consent of the victims. The victims had downloaded the installer from a file-hosting service, under the guise of a freeware program or keys to activate licensed products. This installer downloader the miner’s dropper file to their computer. This installed the Minergate software to the computer, ensuring that it is loaded each time the computer boots and re-installing it if it is deleted.

Often, crypto-miners come with extra services to maintain their presence in the system, launch automatically every time the computer boots and conceal their operation. Such services could, for example try to turn off security software, monitor system activities or ensure that the mining software is always present by restoring it if the files are deleted.

Concealed miners are very difficult to detect because of their specific nature and operating principles. Anyone can choose to install this kind of software and legally use it to mine a crypto-currency.

Monero (XMR) and Zcash are the two currencies most often used in concealed mining. They both ensure the anonymity of transactions – this is clearly very useful for cybercriminals. Even according to conservative estimates, a mining network can generate up to $30,000 per month for its owners.

The above image shows a wallet coded into the miner’s configuration data. At the time of writing, 2,289 XMR had been transferred from this wallet, which at the current exchange rate is equivalent to $208,299.

You can read more here.

Connected hospitals

Technology now reaches into more parts of society than ever before. As a result, organisations that previously didn’t need to think about cyber-security now face cyber-attacks. One example of this is the healthcare industry. Medical information that has traditionally existed in paper form is now to be found in databases, portals and medical equipment.

Data security in medicine is more serious than it seems at first glance. The obvious issue might be the theft and resale of medical data on the black market. However, the possibility of diagnostic data being modified by attackers is even more alarming. Regardless of the goals of the attackers (extortion or attacks targeted at specific patients), there’s a serious risk to patients: after receiving incorrect data, doctors may prescribe the wrong course of treatment. Even if the attempt to substitute data is detected in time, the normal operation of the medical facility may be disrupted, prompting the need to verify all of the information stored on compromised equipment. According to a report by the Centre for Disease Control and Prevention (CDC), the third leading cause of death in the United States comes from medical errors. Establishing a correct diagnosis depends not only on the knowledge and skill of a doctor, but on the correctness of data received from medical devices and stored on medical servers. This makes the resources for connected medicine a more attractive target for attackers. Unfortunately, in some cases, the security of the network infrastructure of healthcare facilities is neglected, and resources that process medical information are accessible from outside sources.

This term ‘connected medicine’ refers to a large number of workstations, servers, and dedicated medical equipment that are connected to the network of a medical institution (a simplified model is shown in the figure below).

Diagnostic devices can be connected to the LAN of an organization or to workstations- for example, through a USB connection. Medical equipment quite often processes data (for example, a patient’s photographs) in DICOM format, an industry standard for images and documents. In order to store them and provide access to them from outside, PACS (Picture Archiving and Communication Systems) are used, which can also be of interest to cybercriminals.

We have put together some recommendations for securing medical facilities. You can find the details here, but the following is a summary of the key points:

  1. Prevent public access to all nodes that process medical data
  2. Assign counter-intuitive names to resources
  3. Periodically update installed software and remove unwanted applications
  4. Don’t connect expensive equipment to the main LAN
  5. Ensure timely detection of malicious activity on the LAN

Using legitimate tools to hide malicious code

Wed, 11/08/2017 - 05:00

The authors of malware use various techniques to circumvent defensive mechanisms and conceal harmful activity. One of them is the practice of hiding malicious code in the context of a trusted process. Typically, malware that uses concealment techniques injects its code into a system process, e.g. explorer.exe. But some samples employ other interesting methods. We’re going to discuss one such type of malware.

Our eye was caught by various samples for .NET that use the trusted application InstallUtil.exe from the Microsoft .NET Framework (information from Microsoft’s website: “The Installer tool is a command-line utility that allows you to install and uninstall server resources by executing the installer components in specified assemblies. This tool works in conjunction with classes in the System.Configuration.Install namespace”).

The technique was described by information security researcher Casey Smith aka subTee (Proof of Concept). Briefly, the console utility InstallUtil.exe runs a malicious .NET assembly, bypassing the entry point of the assembly; all malicious activity is then hidden in the context of the trusted process.

The spreading of malicious samples follows a standard pattern: they basically reach the user in a password-protected archive, and the executable file icons in most cases are chosen specially so that the victim perceives the file as a normal document or photo. We also encountered executable files masquerading as a key generator for common software. To begin with, the malicious content of the generator got inside the %TEMP% folder, where it was run later in the described manner.

Users are misled by executable file icons


All the malicious files we encountered were heavily obfuscated, which complicated their manual analysis. We took the sample 263dc85de7ec717e8940b1ccdd6ee119 and deobfuscated its strings, classes, methods, and fields. Here’s how the file looked before deobfuscation:

Sample before deobfuscation

InstallUtil.exe allows file execution to start not from the .NET assembly entry point: execution begins from a class inherited from System.Configuration.Install.Installer. To facilitate manual analysis, this class was renamed InstallUtilEntryClass in the sample under investigation. The code in static class constructors is known to execute first when the assembly is loaded into memory, a feature utilized by the authors of this piece of malware.

Let’s examine the behavior of the malicious file in the order of methods execution. First up is FirstMainClass, since its constructor is marked with the keyword “static” and assembly execution begins with it:

The static constructor of FirstMainClass that is triggered when the assembly is loaded

The constructor does the following:

  • CheckSandboxieEnvironment() determines whether the file is running in Sandboxie by attempting to load the SbieDll.dll library. If the library can be loaded, the malicious process terminates;
  • CheckVirtualBoxEnvironment() searches for the vboxmrxnp.dll library, which belongs to VitrualBox. If the library can be found, the malicious process likewise terminates;
  • AddResourceResolver() adds a method for handling the resource load event. This method unpacks the assembly, which is packed by the Deflate algorithm, from a specific resource and loads the assembly into memory;

The method responsible for loading the assembly from the resource

The assembly is unpacked from the resource and loaded into memory

  • The UnpackAllAssemblies() method of the AssemblyResourceLoader class iterates through all the assembly resources and, if the resource name contains the string “+||”, unpacks the assemblies from these resources. The assemblies unpacked by this method are required by the malicious file to operate, and are legitimate libraries: Interop.MSScript.Control, Interop.TaskScheduler, SevenZipSharp;
  • RemoveZoneIdentifier() deletes the NTFS alternate stream Zone.Identifier through the command line to prevent a warning at startup if the file was downloaded from the Internet. The authors made a slight mistake in the command line (“cmd.exe /c (echo. > file path:Zone.Identifier) 2 > Null”) by leaving a space between the characters 2 and >, which produces an error in the console:

The warning issued on deleting Zone.Identifier

  • The ElevatePrivilegesProxy() method is the wrapper for the ElevatePrivileges() method, which in turn uses the known UAC bypass technique described by Matt Nelson aka enigma0x3.

Control then passes to the traditional entry point—the Main() method, which is located in the Form5 class:

The traditional entry point is the Main() method

We see that a WMI object is retrieved after a 30-second pause. Next, the ScriptControlClassInstance object is customized, which the language (Visual Basic script) and the body of the script are transferred to:

The script that runs the executable file using InstallUtil.exe

The AddCode() method adds and executes a VB script that runs the current assembly using InstallUtil.exe. After that, the current process is closed by calling Environment.Exit(0).

At the next stage, the malicious object is run using the InstallUtil tool and once more executes the static constructor of the FirstMainClass class examined above; control passes to the static constructor of the InstallUtilEntryClass class, which, as mentioned, is inherited from System.Configuration.Install.Installer:

The static class constructor called by InstallUtil.exe

The functions of this class include:

  • Copying the malicious file to %APPDATA%\program\msexcel.EXE, setting the Hidden+System attributes for the “program” folder, running msexcel.EXE, and terminating the current process;
  • Adding the copied file to autorun (HKEY_LOCAL_MACHINE\Software\Microsoft\Windows\CurrentVersion\Run or HKEY_CURRENT_USER\Software\Microsoft\Windows\CurrentVersion\Run);
  • Creating a task called “filesqmaepq0d.tnk” that runs msexcel.EXE every minute to ensure survival on the victim’s computer;
  • Checking if the malicious process is already running. An event with the name “78759961M” is created, and if such an event already exists in the system, the new process terminates;
  • Creating the Form5 class and calling its destructor.

Let’s sum up the interim results: all the actions described above (entrenchment in the system, elevation of privileges, startup from a trusted application) are essentially laying the foundation for the main task. Let’s move on to analyzing the next stage of the preparatory actions, which will take us closer to the heart of the malicious activity.

The malicious assembly contains, inter alia, five classes inherited from System.Windows.Forms.Form. Inheritance from the Form class is not accidental: in its inheritance hierarchy it implements several interfaces, one of which is IDisposable, which allows to override the Dispose() method for its own purposes. Dispose() methods are called by the garbage collector in order to free up unmanaged resources used by the class when closing or unloading the assembly. Now let’s look at the source code of the Dispose() method of the Form5 class:

The overridden Dispose() method of the Form5 class

As we can see, various methods are executed at each iteration of the cycle, and the results are saved. Let’s take a closer look:

  • At the first iteration, the full path to the RegAsm.exe utility from .NET Framework is retrieved;
  • A chain of nested methods is called with a view to decoding strings from Base64 that are stored in another class and unpacking the resulting array using the SevenZipExtractor library. As a result, we get an array that is the remote administration tool NanoCore Client;
  • The PERun.dll library is loaded from the assembly that was previously unpacked from the resource into memory;
  • A class with the name “RunPE” and the Run method of this class are sought in this library;
  • At the final iteration, the parameters are transferred and the Run method is called.

Knowing that the legalProgramPath variable contains the full path to the legitimate utility RegAsm.exe, PEFileByteArray contains the executable file in the form of a byte array, while the class name is RunPE; it is not hard to figure out that the Run() method employs the technique of hiding malicious code in the address space of the trusted process RunPE. This technique is widely known and described here, for instance.

Deep inside the Run() method, a legitimate utility process is created in CREATE_SUSPENDED state (the sixth parameter is 4u):

Creating a legitimate program process in CREATE_SUSPENDED state

Eventually, the RegAsm.exe process is loaded in the address space and starts to execute the payload: the remote administration tool NanoCore Client. Only trusted processes remain in the list of running processes, and even an experienced user might not realize that the system is compromised:

Only legitimate utilities can be seen in the list of running processes

RegAsm.exe was chosen as the “carrier” because (a) it is a legitimate utility from Microsoft, (b) it is located in the same directory as InstallUtil.exe, and (c) a utility from .NET Framework calling another utility from the same framework is less suspicious than calling, say, notepad.exe. In fact, the use of RegAsm.exe is not critical: the “carrier” could be any program that does not arouse the suspicion of security software and users. It is also important that all actions involving a malicious module are executed in memory, which allows file scanners to be bypassed.

As we’ve mentioned, this sample contains NanoCore Client, which can be used to control the victim’s computer, take screenshots, record keystrokes, download files, and much more. It should be noted that the payload here can be anything: from “fashionable” encrypters and miners to advanced Trojans.


Malware writers employ various tricks to conceal malicious activity, and the above technique allowing the execution of malicious code in the context of two legitimate programs is an obvious example. Detecting this kind of concealment method requires a behavioral analysis of the program. Kaspersky Lab’s security solutions detect this behavior as PDM: Trojan.Win32.Generic and PDM: Exploit.Win32.Generic.


263DC85DE7EC717E8940B1CCDD6EE119 payload: EF8AF3D457DBE875FF4E3982B34F1DE9
3E4825AA1C09E27C2E6A1309BE8D6382 payload: 82709B139634D74DED404A516B7952F0
7E3863F827C1696835A49B8FD7C02D96 payload: D1A9879FFCB14DF70A430E59BFF5EF0B
8CB8F81ECF1D4CE46E5E96C866939197 payload: D8652841C19D619D2E3B5D7F78827B6E
FDF4086A806826503D5D332077D47187 payload: BF4A3F4B31E68B3DE4FB1F046253F2D0

DDoS attacks in Q3 2017

Mon, 11/06/2017 - 05:00

News Overview

In the third quarter of 2017, the trends of the preceding quarters continued to develop further. The number of DDoS attacks in China, the United States, South Korea and Russia increased, which were reflected in the statistics we gathered for botnets. A sharp surge in the number (more than 450 daily) and power (up to 15.8 million packets per second) of attacks was registered in the ‘Australian sector’. The cost of protection increased accordingly: for example, in early September, six IB vendors entered into a $50 million contract with the Singapore government (the previous three-year contract cost the state half that amount).

The biggest success in combating DDoS attacks was the taking down of the huge (hundreds of thousands of devices in more than a hundred countries) WireX botnet. The botnet had been secretly working on Android devices and proliferating via legitimate Google Play applications. The joint actions of Google, Samsung and several large IT security vendors were required to take down the botnet. Given the deplorable state of security on the Internet of things and in micro-applications, such findings are now likely to occur on a fairly regular basis.

Cybercriminals are using their brains as well as their brawn. In mid-August, Imperva described Pulse Wave technology capable of increasing the power of a DDoS attack thanks to a vulnerability in hybrid and cloud technologies. The analysts at Imperva believe that most DDoS attacks will soon follow a similar pattern: short but powerful sudden “punctuated” attacks that last for several hours or several days.

The targets within the scope of the cybercriminals’ interest remain the same. In the political arena, the increase in the number of attacks has even triggered a process of qualitative change: some are voicing the belief that DDoS attacks are a legitimate form of democratic protest. However, the effectiveness of this method is still questionable: the two most notable political acts of the third quarter (an attack on the DreamHost hosting provider and on a libertarian site) achieved nothing apart from greater publicity for the attacked resources.

Cases of blackmail involving DDoS attacks – or rather, attempts that aren’t always very well executed –have become more frequent. While in the previous quarter companies preferred to pay off the attackers, mass mailings with threats are now often perceived as just another wave of spam.

As a means of applying pressure, DDoS attacks are still more beneficial in industries where downtime and communication failures lead to lost profits and reputation. The gaming industry is becoming even more attractive for cybercriminals: the profits here are estimated in the hundreds of billions of dollars, while security is still far from perfect, with hybrid gaming platforms vulnerable to attacks via the links between resources and applications.

In Q3, there were three high-profile incidents involving gaming platforms (not including the DDoS attack on Final Fantasy’s servers, which, according to Square Enix, began in June and lasted till the end of July).

Firstly, in mid-August, Blizzard Entertainment reported a flood of junk traffic that caused problems for players of Overwatch and World of Warcraft.

Secondly, at the beginning of September, the Americas Cardroom online poker site began to experience difficulties. The attack (not the first to target the resource) followed the notorious pattern “demonstrate force, demand a ransom”. The site’s management refused to pay, but was forced to cancel – or more precisely, to delay – a poker championship that was already under way.

At the end of the quarter, on 30 September, the site of the UK National Lottery was seriously affected: for 90 minutes players were unable to place their stakes online or via applications, which caused the service serious losses.

It appears that constant DDoS attacks on the entertainment industry is becoming the new normal: the largest companies will either have to seriously reconsider their approach to security or put customer loyalty at risk. Some of them have started eliminating possible vectors on their own. For example, Netflix (yet another entertainment platform that could lose customers due to a loss of communication) found a serious vulnerability in API and developed two tools to deal with the infected applications.

Probably the most curious attack of the quarter was also related to the entertainment and gaming industry: the cybercriminals hacked a US casino via a smart fish tank. It had nothing to do with DDoS attacks, but it’s interesting that criminals managed to break through to the mainframe and steal 100 GB of confidential data from the organization, although the fish tank was installed on its own VPN. It is highly likely that in the near future the entertainment and gaming sector will be on a par with the financial sector when it comes to the scope and ingenuity of large-scale attacks.

Quarter Trends

In term of trends, there was a fairly new vector of attacks related to the now notorious crypto- currencies. More and more attacks are targeting Initial Coin Offering (ICO) platforms – a type of crowdfunding. Since blockchain technology allows transactions to be conducted safely, ICOs are quickly gaining in popularity. But there are risks as well: with the rapid growth and the increasing turnover of crypto-currencies, such platforms are subjected to cyberattacks, including DDoS attacks. The broad availability of the platform guarantees reliable and secure transactions, while DDoS attacks are aimed at breaking the operability of the service and thus discrediting it or, even worse, creating a smokescreen for more sophisticated types of attacks.

Another detail of this quarter is the increase in the proportion of mixed, multi-component (SYN + TCP Connect + HTTP-flood + UDP flood) attacks. As forecasted earlier, they are gradually gaining in popularity. There is nothing fundamentally new in these attacks, but in the right hands they can be quite effective.

Statistics for botnet-assisted DDoS attacks Methodology

Kaspersky Lab has extensive experience of combating cyber threats, including DDoS attacks of various complexity types and ranges. The experts of the company have been tracking the actions of botnets by using the DDoS Intelligence system.

Being part of the Kaspersky DDoS Prevention solution, the DDoS Intelligence system is intended to intercept and analyze commands sent to bots from command-and-control servers and requires neither infecting any user devices nor the actual execution of cybercriminals’ commands.

This report contains DDoS Intelligence statistics for the third quarter of 2017.

In the context of this report, it is assumed that an incident is a separate (single) DDoS-attack if the interval between botnet activity periods does not exceed 24 hours. For example, if the same web resource was attacked by the same botnet with an interval of 24 hours or more, then this incident is considered as two attacks. Also, bot requests originating from different botnets but directed at one resource count as separate attacks.

The geographical locations of DDoS-attack victims and C&C servers that were used to send commands are determined by their respective IP addresses. The number of unique targets of DDoS attacks in this report is counted by the number of unique IP addresses in the quarterly statistics.

It is important to note that DDoS Intelligence statistics are limited only to those botnets that have been detected and analyzed by Kaspersky Lab. It should also be noted that botnets are just one of the tools for performing DDoS attacks; thus, the data presented in this report do not cover every single DDoS attack occurred during the indicated period.

Q3 summary
  • Resources in 98 countries were attacked in Q3 2017 vs. 86 in Q2 2017.
  • As in Q2, around half of all attacks (51.56%) originated in China.
  • China, the US, and South Korea remained leaders in terms of both number of attacks and number of targets. According to the number of reported C&C servers, the same countries are make up the TOP 3, though South Korea calimed first place this time.
  • The longest DDoS attack was 215 hours, a decrease of 28% compared to Q2. At the same time, the share of attacks that lasted less than 50 hours remained practically unchanged (99.6% in Q3 vs. 99.7% in Q2).
  • As in the previous quarter, there was a considerable drop in the proportion of attacks over TCP (down to 11.2% from 28.2%) and ICPM (down to 7.1% from 9.42%). This caused a rise in the percentage of SYN floods and HTTP attacks.
  • The proportion of Linux botnets continued to grow. Such botnets were responsible for 69.62% of attacks in Q3 compared to 51.23% in Q2.
Geography of attacks

DDoS attacks were registered in 98 countries in Q3, where the largest number of the attacks were aimed at China (63.30% of all attacks), which is 5.3 p.p. higher than the previous quarter. South Korea’s share fell from 14.17% to 8.70%, moving it to third place. The US came second despite the percentage of attacks originating from this country falling from 14.03% to 12.98%.

The top 10 accounted for 93.56% of all attacks. Germany (1.24%) re-entered the top 10, replacing Italy out of the rating. Hong Kong (1.31%) dropped from 4th to 7th, having lost 1.07 p.p. Russia (1.58%) gained 0.35 p.p. and was once again in fourth place. The UK remained fifth while the Netherlands saw its share go up from 0.84% to 1.31%, moving it to sixth.

Distribution of DDoS attacks by country, Q2 2017 vs. Q3 2017

91.27% of all attacks were aimed at targets in the countries of the top 10 in Q3 2017.

Distribution of unique DDoS-attack targets by country, Q2 2017 vs. Q3 2017

China remained in first place: 51.56% of all targets were located in the territory of the country, an increase of 4.14 p.p. compared to Q2. At the same time, the US and South Korea remained second and third respectively, although the proportion of targets in the territories of both countries fell considerably: from 18.63% to 17.33% in the US, and from 16.35% to 11.11% in South Korea.

The share of targets located in the territory of Russia grew from 1.33% in Q2 to 2.24% in Q3, which saw Russia move up from seventh to fourth place. Australia and Italy left the top 10 and were replaced by France (1.43%) and Germany (1.65%).

Dynamics of the number of DDoS attacks

The number of attacks per day ranged from 296 (24 July) to 1508 (26 September) in Q3 2017. The peak numbers were registered on 27 July (1399) and 24 September (1497). A relative downturn was registered on 28 July (300), 31 May (240), and 25 September (297).

Dynamics of the number of DDoS attacks in Q3 2017*
*Since DDoS attacks may continuously last for several days, one attack may be counted several times in the timeline, i.e., once per day.

In Q3 2017, Monday remained the quietest day for DDoS attacks (10.39% vs 11.78% in the previous quarter), while Thursday became the busiest day (17.54%). Last quarter’s leader, Saturday, came second (15.59%) followed by Sunday (14.89%) and Tuesday (14.79%).

Distribution of DDoS attacks by day of the week, Q2 vs Q3 2017

Types and duration of DDoS attacks

As in the previous quarter, the number of SYN DDoS attacks continued to grow, rising from 53.26% to 60.43% in Q3 2017. At the same time, the percentage of TCP DDoS attacks plummeted from 18.18% to 11.19%, which did not affect second position in the rating for this type of attack. Both UDP and ICMP attacks became quite rare: their share dropped from 11.91% to 10.15% and from 9.38% to 7.08% respectively. Meanwhile, the popularity of HTTP attacks increased from 7.27% to 11.6%, which placed them in third.

Distribution of DDoS attacks by type, Q3 2017

The number of long-term attacks remained almost unchanged from the previous quarter: 0.02% of attacks lasted more than 150 hours (vs 0.01%). The longest attack lasted for 215 hours, 62 hours shorter than the record in Q2. At the same time, the share of attacks that lasted 4 hours or less dropped from 85.93% in Q2 to 76.09% in Q3. Thus, the percentage of attacks lasting from 5 to 49 and from 50 to 99 hours increased, accounting for 23.55% and 0.3% of all attacks respectively.

Distribution of DDoS attacks by duration (hours), Q2 vs Q3 2017

C&C servers and botnet types

The top 3 countries with the greatest number of detected C&C servers remained unchanged from Q2: South Korea, whose share grew from 49.11% to 50.16%, remained top. The US retained second place (16.94% vs 16.07% in Q2). China remained third although its share dropped from 7.74% to 5.86%. The top 3 countries accounted for 72.96% of C&C servers in total, which is only slightly more than in the previous quarter.

The top 10 included Italy (1.63%) and the UK (0.98%), which ousted Canada and Germany in Q3. Compared to Q2 2017, there was a significant increase in the shares of France (up to 2.93% from 1.79%) and Russia (up to 3.58% from 2.68%).

Distribution of botnet C&C servers by country in Q3 2017

In Q3, Linux-based botnets continued to win back positions from Windows: the share of detected Linux-based botnets comprised 69.62%, while the percentage of Windows-based botnets dropped to 30.38%.

Correlation between Windows- and Linux-based botnet attacks, Q3 2017


In the third quarter of 2017, we registered a considerable increase in the number of both DDoS attacks and their targets. Traditionally, China is the country with the largest number of attack sources and targets. It was followed by the United States and South Korea. The popularity of Windows OS as a basis for creating a botnet has fallen noticeably, while the share of Linux-based botnets increased proportionally.

Among this quarter’s trends were increased attacks on ICO platforms: in Q3, crypto-currency was widely discussed both on the Internet and in the mass media, and cybercriminals did not ignore its popularity. Yet another detail of this quarter is the growth in the proportion of multi-component attacks, consisting of various combinations of SYN, TCP Connect, HTTP flood and UDP flood techniques.

Spam and phishing in Q3 2017

Fri, 11/03/2017 - 06:00

Quarterly highlights Blockchain and spam

Cryptocurrencies have been a regular theme in the media for several years now. Financial analysts predict a great future for them, various governments are thinking about launching their own currencies, and graphics cards are swept off the shelves as soon as they go on sale. Of course, spammers could not resist the topics of cryptocurrency, mining and blockchain technology.

Last quarter we wrote that many Trojans were downloading ‘miners’ as a payload on victims’ computers, and in third quarter of 2017 this practice became even more widespread.

Fraud, cryptocurrencies and binary options

Financial fraud makes very active use of the cryptocurrency topic: users receive messages that vividly describe the use of special software for trading on the cryptocurrency market and how it can secure their financial future.

Examples of emails with offers “to secure your financial future”

After clicking on a link, users end up on a site where they are once again persuaded to join the ranks of the rich who only have one problem in life – how to spend their money. In reality, such sites are partners for shady brokerage houses, and purveyors of new, inexperienced customers. It is there that new users are redirected.

The plan is to get the victim to deposit a certain amount to their account, usually several hundred dollars, for the opportunity to start trading. We should note here that we’re no longer talking about cryptocurrencies – in most cases, trading involves binary options.

The problem is not even in the questionable legality of the actual trading, but that no one guarantees the honesty of the brokerage offices and, consequently, there are no guarantees that the invested funds will be returned. The fraudsters start by motivating people to invest more and more money, and then simply disappear, leaving the victim to read angry reviews on the Internet from other cheated depositors.

There are also more primitive types of fraud, where the email directly asks the recipient to transfer bitcoins to a specific wallet, with a promise to return the investment with interest five days later. But only the most naïve recipients are likely to fall for such an offer.

Naive users are invited to “invest” bitcoins for a short time at a high high rate of interest


Another example of the cryptocurrency theme being used in spam is that of webcasts. In most cases, scammers suggest taking a study course that will help the user understand more about cryptocurrencies and how to invest in them. Of course, the sums invested in “training” will result in huge profits in the near future, according to the organizers.

Natural disasters and the ‘White House administration’

In August and September, the world’s attention was focused on hurricanes Irma and Harvey, and the earthquake in Mexico. There were dozens of victims of these disasters, and the damage caused was estimated to be billions of dollars. These tragic events inevitably attracted the attention of so-called Nigerian scammers trying to cash in on people’s grief. They sent messages on behalf of family members whose relatives died during the hurricanes and asked for help obtaining an inheritance left by them. Natural disasters were also mentioned in emails promoting job offers and loans.

In the third quarter, ‘Nigerian’ letters also mentioned the name of Donald Trump, the current US president. The authors pretended to be representatives of state or banking organizations, and to make their message sound more important they claimed they were appointed by the US president or were acting on his behalf. The spammers spun the standard tales in their fraudulent letters, promising millions of dollars to users, with the scammers asking for personal information so that they could supposedly track the money transfer. The letters contained identical text but with different layouts and contact details.

Letters ‘from the US president’s office’

B2B fakes in malicious emails

There is still a tendency to create emails with malicious attachments for fake commercial offers. At times their quality is so good that you suspect they could be a man-in-the-middle attack.

The file in the attachment is detected as HEUR: Trojan.Java.Agent.gen. This malware is written to startup and tries to close programs such as Process Hacker, system explorer and security software processes. It then communicates with the remote server and waits for the command to install other malicious programs

The attachment is detected as HEUR: Exploit.MSOffice.Generic, exploiting the vulnerability CVE-2017-0199 in MS Word. As a result, other malicious programs are downloaded to the victim’s computer

Both archives contain the same malicious object, detected as Trojan.Win32.VBKrypt.xtgt. It collects information from the victim’s computer and transfers it to the remote server

Release of new iPhone

In September, Apple unveiled the new models of its smartphone – iPhone 8 and iPhone X. This event was widely covered in the media, and spammers, weren’t going to miss out.

Even before the official presentation, we began to record spam mailings with offers to test the updated phone for free and participate in a prize draw to win one. Some mailings even reported the recipient had won a device before it was publicly unveiled. In most cases, the links in these emails could end up downloading Reimage Repair ‘advertising software’. Immediately after the release of the smartphone, Chinese factories got in on the act, sending out emails advertising various accessories for the new model. Our traps also recorded a large volume of phishing associated with the purchase and delivery of the popular gadget.

Statistics Proportion of spam in email traffic

Percentage of spam in global email traffic, Q2 and Q3 2017

In the third quarter of 2017, the largest share of spam was recorded in September – 59.56%. The average share of spam in global email traffic was 58.02%, which was almost 1.05 p.p. more than the average for the previous quarter.

Sources of spam by country

Sources of spam by country, Q3 2017

According to the results for the third quarter of 2017, China (12.24%) became the biggest source of spam, after finishing third the previous quarter. Last quarter’s leader Vietnam (11.17%) was second after a decrease of 1.2 p.p. The US fell one place to third (9.62%), while India (8.49%) remained fourth in this rating. Iran rounded off the top 10, accounting for 2.07% of all spam.

Spam email size

Breakdown of spam emails by size, Q2 and Q3 2017

The share of very small emails (up to 2 KB) in spam increased by 9.46 p.p. to 46.87% in the third quarter. The proportion of emails between 5 and 10 KB in size also increased by 6.66 p.p. compared with the previous quarter and amounted to 12.6%.

The number of emails between 10 and 20 KB decreased, however, with their share falling by 7 p.p. There was also a decrease in emails sized 20 to 50 KB. Their share this quarter amounted to 19%, which was a fall of 8.16 p.p. compared to the previous reporting period.

Overall, the number of very small emails continues to grow.

Malicious attachments in email Top 10 malware families

TOP 10 malware families in Q3 2017

Backdoor.Java.QRat (3.11%) became the most widespread malicious program family in email traffic. Next came the Trojan-Downloader.VBS.Agent family (2.95%), followed by Trojan-Downloader.JS.SLoad (2.94%). The newcomers in this rating – Trojan.Win32.VBKrypt and Trojan-Downloader.VBS.SLoad (a VBS script that downloads and launches other malicious programs on the victim machine, usually cryptographers) occupy fifth and eighth places with 2.64% and 2.02% respectively. The Trojan.PDF.Badur family (1.79%) rounds off the top 10.

Countries targeted by malicious mailshots

Distribution of email antivirus verdicts by country, Q3 2017

Germany remained the country targeted most by malicious mailshots in the third quarter of 2017. Its share increased by 6.67 p.p. and amounted to 19.38%.

China came second, with 10.62% of mail antivirus verdicts recorded there – a drop of 1.47 p.p. compared to Q2. Russia, which came fifth the previous quarter, completed the top three (9.97%) after its share increased by 4.3 p.p. Fourth and fifth were occupied by Japan (5.44%) and Italy (3.90%) respectively.


In the third quarter of 2017, the anti-phishing system prevented 59,569,508 attempted visits to phishing pages on the computers of Kaspersky Lab users. Overall, 9.49% of unique users of Kaspersky Lab products worldwide were attacked by phishers in Q3 2017.

Geography of attacks

The country with the largest percentage of users affected by phishing attacks was once again Brazil (19.95%, +1.86p.p.).

Geography of phishing attacks*, Q3 2017
* Number of users on whose computers the Anti-Phishing system was triggered as a percentage of the total number of Kaspersky Lab users in that country

Australia (16.51%) came second after its share increased by 3.81 p.p. In third place was New Zealand (15.61%, + 3.55pp). China (12.66%) fell from second place to fourth, with its share losing 0.19 p.p. Next came France (12.42%), Peru (11.73%), Argentina (11.43%), Canada (11.14%), Qatar (10.51%,) and Georgia (10.34%).

Brazil 19.95% Australia 16.51% New Zealand 15.61% China 12.66% France 12.42% Peru 11.73% Argentina 11.43% Canada 11.14% Qatar 10.51% Georgia 10.34%

TOP 10 countries by percentage of users attacked

Organizations under attack Rating the categories of organizations attacked by phishers

The rating of attacks by phishers on different categories of organizations is based on detections of Kaspersky Lab’s heuristic anti-phishing component. It is activated every time a user attempts to open a phishing page while information about it has not yet been included in Kaspersky Lab’s databases. It does not matter how the user attempts to open the page – by clicking a link in a phishing email or in a message on a social network or, for example, as a result of malware activity. After the security system is activated, a banner is displayed in the browser warning the user about a potential threat.

In the third quarter of 2017, almost half (47.54%) of the heuristic components of the anti-phishing system were recorded on pages with references to brands from financial categories such as Banks (24.1%, + 0.61 p.p.), Payment systems (13.94%, -4.46 p.p.) and Online stores (9.49%, -0.08 p.p.).

Distribution of organizations affected by phishing attacks by category, Q3 2017

Hot topics this quarter Airline tickets

Last quarter we described a scam involving a free giveaway of airline tickets supposedly by popular airlines, with information being spread via reposts from victims on a social network. In the third quarter, scammers continued to spread the ‘giveaway’ using WhatsApp instead. Judging by the decrease in the number of anti-phishing verdicts in the Airlines category, however, we can assume that this approach wasn’t as effective.

The downturn may also be due to the fact that scammers switched to ‘prize draws’ not only for air tickets but also other prizes, for example, sports shoes, cinema tickets, gift cards for Starbucks, etc.

Before you could claim your prize you had to share information about the prize draw with eight contacts on WhatsApp.

After clicking the button, users are redirected to WhatsApp.

The redirect function in the instant messenger and the message that has to be sent to contacts

This is what the message looks like in the app

The message needs to be sent a minimum of eight times

After sending the message to their contacts the victim, instead of winning a prize, is redirected to some dubious resource, for example, a page where malicious extensions are installed, a new survey, etc.


WhatsApp users are also subjected to phishing attacks that hide behind the app brand.

More often than not the scammers try to steal money on the pretext of updating the application or paying for a subscription. At one time WhatsApp really did request a subscription payment, although now it’s free.

Scammers offer a choice of subscription – for one year, three years or five. However, victims will lose much more than the stated amount if they enter their bank card details on such a site.


Netflix users are another popular target of phishers. The number of attacks on them increased in the third quarter. The criminals usually coax bank card details from users on the pretext of a failed payment or other problems linked to subscription renewal.

Green Card

On the eve of the Green Card lottery conducted by the US government in October-November of each year, we are seeing a surge in activity by scammers offering help to apply.

After completing the form on the fraudulent site, the user is asked to pay for their application. If the victim enters their bank card details, much more money than the amount indicated on the site can end up being withdrawn from their account.

Rap battle

Even niche events can be good cover for phishing activity. On 15 October, a rap battle was held between Russian artist Oxxxymiron and Dizaster, one of the best battle MCs in the US. This followed another battle that took place just a few months earlier between Oxxxymiron and Slava KPSS. Less than 12 hours later a video of the event had gained around 5 million views – and it wasn’t just thematic sites writing about the battle but also a lot of the mainstream Russian media.

Shortly before the publication of the official video, phishing web pages dedicated to the event began to appear online:

If a user tried to view the video, they were prompted to first sign in to the popular Russian social network VKontakte.

After entering the login and password, the victim was redirected to the official page of the Versus site on the social network, and their personal data went to the scammers.

TOP 3 attacked organizations

Fraudsters continue to focus most of their attention on the most popular brands, enhancing their chances of a successful phishing attack. More than half of all detections by Kaspersky Lab’s heuristic anti-phishing component are for phishing pages hiding behind the names of fewer than 15 companies. At the same time, the composition of the top three has remained unchanged for several quarters:

Organization % of detected phishing links Facebook 7.96 Microsoft Corporation 7.79 Yahoo! 4.79 Conclusion

In terms of the average share of spam in global email traffic (58.02%), the third quarter of 2017 was almost identical to the previous reporting period: once again growth was slightly more than one percentage point – 1.05 (and 1.07 p.p. in Q2 2017). As in previous quarters, spammers were quick to react to high-profile events and adapted their fraudulent emails to the news agenda. This quarter they were quickly to use the theme of natural disasters following hurricanes Irma and Harvey, and the earthquake in Mexico. The popular theme of cryptocurrency was also used: trusting victims were offered seminars and ‘help’ with trading that came with profits guaranteed.

Scammers continued to use all available communication channels to spread phishing content, including social networks and instant messengers: in the current quarter, the anti-phishing component prevented more than 59 million attempts to redirect to phishing pages, which is 13 million more than in Q2.

The most common malware family in the third quarter of 2017 was Backdoor.Java.QRat (3.11%), followed by Trojan-Downloader.VBS.Agent (2.95%), and Trojan-Downloader.JS.SLoad (2.94%).

Silence – a new Trojan attacking financial organizations

Wed, 11/01/2017 - 07:26

More information about the Silence Trojan is available to customers of Kaspersky Intelligence Reporting Service. Contact:

In September 2017, we discovered a new targeted attack on financial institutions. Victims are mostly Russian banks but we also found infected organizations in Malaysia and Armenia. The attackers were using a known but still very effective technique for cybercriminals looking to make money: gaining persistent access to an internal banking network for a long period of time, making video recordings of the day to day activity on bank employees’ PCs, learning how things works in their target banks, what software is being used, and then using that knowledge to steal as much money as possible when ready.

We saw that technique before in Carbanak, and other similar cases worldwide. The infection vector is a spear-phishing email with a malicious attachment. An interesting point in the Silence attack is that the cybercriminals had already compromised banking infrastructure in order to send their spear-phishing emails from the addresses of real bank employees and look as unsuspicious as possible to future victims.

The attacks are currently still ongoing.

Technical details

The cybercriminals using Silence send spear-phishing emails as initial infection vectors, often using the addresses of employees of an already infected financial institution, with a request to open an account in the attacked bank. The message looks like a routine request. Using this social engineering trick, it looks unsuspicious to the receiver:

Spear-phishing email in Russian.

Malicious .chm attachment md5 dde658eb388512ee9f4f31f0f027a7df Type Windows help .chm file

The attachment we detected in this new wave is a “Microsoft Compiled HTML Help” file. This is a Microsoft proprietary online help format that consists of a collection of HTML pages, indexing and other navigation tools. These files are compressed and deployed in a binary format with the .CHM (compiled HTML) extension. These files are highly interactive and can run a series of technologies including JavaScript, which can redirect a victim towards an external URL after simply opening the CHM. Attackers began exploiting CHM files to automatically run malicious payloads once the file is accessed. Once the attachment is opened by the victim, the embedded .htm content file (“start.htm”) is executed. This file contains JavaScript, and its goal is to download and execute another stage from a hardcoded URL:

Part of start.htm embedded file

The goal of the script is to download and execute an obfuscated .VBS script which again downloads and executes the final dropper

Obfuscated VBS script that downloads binary dropper

Dropper md5 404D69C8B74D375522B9AFE90072A1F4 Compilation Thu Oct 12 02:53:12 2017 Type Win32 executable

The dropper is a win32 executable binary file, and its main goal is to communicate with the command and control (C&C) server, send the ID of the infected machine and download and execute malicious payloads.

After executing, the dropper connects to the C&C using a GET request, sends the generated victim ID, downloads the payloads and executes them using the CreateProcess function.

C&C connect request string with ID

C&C connect procedure


The payloads are a number of modules executed on the infected system for various tasks like screen recording, data uploading etc.

All the payload modules we were able to identify are registered as Windows services.

Monitoring and control module md5 242b471bae5ef9b4de8019781e553b85 Compilation Tue Jul 19 15:35:17 2016 Type Windows service executable

The main task for this module is to monitor the activity of the victim. In order to do so it takes multiple screenshots of the victim´s active screen, providing a real-time pseudo-video stream with all the victim´s activity. A very similar technique was used in the Carbanak case, where this monitoring was used to understand the victim´s day to day activity.

The module is registered and started by a Windows service named “Default monitor”.

Malicious service module name

After the initial startup, it creates a Windows named pipe with a hardcoded value – “\\.\pipe\{73F7975A-A4A2-4AB6-9121-AECAE68AABBB}”. This pipe is used for sharing data in malicious inter-process communications between modules.

Named pipe creation

The malware decrypts a block of data and saves it as a binary file with the hardcoded name “mss.exe” in a Windows temporary location, and later executes it using the CreateProcessAsUserA function. This dropped binary is the module responsible for the real-time screen activity recording.

Then, the monitoring module waits for a new dropped module to start in order to share the recorded data with other modules using the named pipe.

Screen activity gathering module md5 242b471bae5ef9b4de8019781e553b85 Compilation Tue Jul 19 15:35:17 2016 Type Windows 32 executable

This module uses both the Windows Graphics Device Interface (GDI) and the Windows API to record victim screen activity. This is done using the CreateCompatibleBitmap and GdipCreateBitmapFromHBITMAP functions. Then the module connects to the named pipe created by the previously described module and writes the data in there. This technique allows for the creation of a pseudo-video stream of the victim’s activity by putting together all the collected bitmaps.

Writing bitmaps to pipe

C&C communication module with console backconnect

md5 6A246FA30BC8CD092DE3806AE3D7FC49 Compilation Thu Jun 08 03:28:44 2017 Type Windows service executable

The C&C communication module is a Windows service, as are all the other modules. Its main functionality is to provide backconnect access to the victim machine using console command execution. After the service initialization, it decrypts the needed Windows API function names, loads them with LoadLibrary and resolves with GetProcAddress functions.

WinAPI resolving

After successful loading of the WinAPI functions, the malware tries to connect to the C&C server using a hardcoded IP address (185.161.209[.]81).


The malware sends a special request to the command server with its ID and then waits for a response, which consists of a string providing the code of what operation to execute. The options are:

  • “htrjyytrn” which is the transliteration of “reconnect” (“реконнект” in russian layout).
  • “htcnfhn” which is the transliteration of “restart” (“рестарт” in russian layout).
  • “ytnpflfybq” which is the transliteration of “нет заданий” meaning “no tasks”

Finally the malware receives instructions on what console commands to execute, which it does using a new cmd.exe process with a parameter command.

Instruction check

The described procedure allows attackers to install any other malicious modules. That can be easily done using the “sc create” console command.

Winexecsvc tool

md5 0B67E662D2FD348B5360ECAC6943D69C Compilation Wed May 18 03:58:26 Type Windows 64 executable

Also, on some infected computers we found a tool called the Winexesvc tool. This tool basically provides the same functionality as the well-known “psexec” tool. The main difference is that the Winexesvc tool enables the execution of remote commands from Linux-based operating system. When the Linux binary “winexe” is run against a Windows server, the winexesvc.exe executable is created and installed as a service.


Attacks on financial organization remain a very effective way for cybercriminals to make money. The analysis of this case provides us with a new Trojan, apparently being used in multiple international locations, which suggests it is an expanding activity of the group. The Trojan provides monitoring capabilities similar to the ones used by the Carbanak group.

The group uses legitimate administration tools to fly under the radar in their post-exploitation phase, which makes detection of malicious activity, as well as attribution more complicated. This kind of attack has become widespread in recent years, which is a very worrisome trend as it demonstrates that criminals are successful in their attacks. We will continue monitoring the activity for this new campaign.

The spear-phishing infection vector is still the most popular way to initiate targeted campaigns. When used with already compromised infrastructure, and combined with .chm attachments, it seems to be a really effective way of spreading, at least among financial organizations.


The effective way of protection from targeted attacks focused on financial organizations are preventive advanced detection capabilities such as a solution that can detect all types of anomalies and scrutinize suspicious files at a deeper level, be present on users’ systems. The Kaspersky Anti Targeted Attack solution (KATA) matches events coming from different infrastructure levels, discerns anomalies and aggregates them into incidents, while also studying related artifacts in a safe environment of a sandbox. As with most Kaspersky products, KATA is powered by HuMachine Intelligence, which is backed by on premise and in lab-running machine learning processes coupled with real-time analyst expertise and our understanding of threat intelligence big data.

The best way to prevent attackers from finding and leveraging security holes, is to eliminate the holes altogether, including those involving improper system configurations or errors in proprietary applications. For this, Kaspersky Penetration Testing and Application Security Assessment services can become a convenient and highly effective solution, providing not only data on found vulnerabilities, but also advising on how to fix it, further strengthening corporate security.


Kaspersky lab products detects the Silence trojan with the following verdicts:

Full IOC’s and YARA rules delivered with private report subscription.


Tales from the blockchain

Tue, 10/31/2017 - 05:00

Cryptocurrency has gradually evolved from an element of a new world, utopian economy to a business that has affected even those sectors of society least involved in information technology. At the same time, it has acquired a fair number of “undesirable” supporters who aim to enrich themselves at the expense of other users: attackers who release miners embedded in user JS scripts, or plan to implement miners into IoT devices at the production stage; hidden in countless variations of Trojans in conjunction with SMB exploits etc.

We will tell you two unusual success stories that happened on the “miner front”. The first story echoes the TinyNuke event and, in many respects gives an idea of the situation with miners. The second one proves that to get crypto-currency, you don’t need to “burn” the processor.

DiscordiaMiner and fights on forums

In early June, our analysts found a new and seemingly unremarkable Trojan that unloaded the miner of the popular Montero crypto-currency. However, in the course of further research, we uncovered many interesting details that we would like to share with you.

Kaspersky Lab products detect this Trojan as Trojan.Win32.DiscordiaMiner. It works as follows:

  1. Creates a number of directories in the system to download the necessary files;
  2. Copies itself in C:\ProgramData\MicrosoftCorporation\Windows\SystemData\Isass.exe;
  3. Gets the update from the server;
  4. Creates an autorun task;
  5. Gets the miner files;
  6. Gets the credentials of the user in whose name it wants to run the mining;
  7. Starts the miner.

All interaction with the command server (C&C) occurs in the open, with the help of GET requests, without any check or verification. In all samples, the hxxp://api[.]boosting[.]online address is provided as the C&C. The line associated with the individual user (etc. MTn31JMWIT) and the address of the required resource – the list of files, the update, etc. – are added to the server address. Example: hxxp://api[.]boosting[.]online/MTn31JMWIT/getDiscordia

Discord on the forum

As mentioned above, at a certain point in its work, the Trojan is instructed to issue a command to run the miner: it specifies the email of the user who has “done the job”. It looks like this:

-user <user_email> -xmr

Using the value of the <user_email> argument, with the first line of the search results we get the Trojan-related topic on the Russian-language forum:

On this forum thread there is a wide discussion of the Trojan’s work details. The most interesting part of the discussion is on page 21 – the forum participants accuse the Trojan’s author of substituting users’ addresses with his own. Among other things, there is also a dialogue on the chat app, Telegram where the author explains this substitution as a banal mistake.

On the forum, the author of DiscordiaMiner references the short lifespan of this error as an argument in his defense:

He also mentions the figure of 200,000 infected machines. It is difficult to say how true this is. However, in the malware samples we received, the email that the “prosecutor” refers to is often named. Examples of other addresses: ilya-soro*****, v*****, topne*****, J ***** m @, steamfa*****, me*****, x*****, piedmont ***** lines @ yahoo. com.

Among other things, in the course of the dispute the author mentions that the source codes of the Trojan DiscordiaMiner are now publicly available.

Indeed, the first line of the search results provides the link to the author’s repository.

In addition to the source codes, which really do coincide fully with the restored Trojan code, the repository also includes very informative diagrams of the Trojan’s operation, the samples of documents used for distribution as well as instructions for how exactly the UAC is to be bypassed. The pictures below are taken from the repository (which is currently unavailable).

The source codes are presented in full and, apparently, only the user-associated string (ClientID) varies from assembly to assembly.

Although the “dumping” of program source code is not unique, this case in many respects echoes the NukeBot story – the same disputes on a forum followed by the publication of the source codes by the author with the aim of “protecting honor and dignity”. Another common feature is the “minimalistic” design of both Trojans: NukeBot could only embed web-based injections into the browser, while DiscordiaMiner can download and run files from a remote server. But we cannot say whether these two bots have any more specific connections.




It’s extremely rare for authors of mining software to become fabulously wealthy. With a few exceptions, the wallets used by attackers contain a total of $50-100, received from all incoming transfers during the entire period of the Trojan’s work. However, there are those that do not go down the beaten path, and benefit from “alternative” ways. The authors of the CryptoShuffler Trojan belong in this category.

Kaspersky Lab products detect this Trojan as Trojan-Banker.Win32.CryptoShuffler.gen. MD5 of the file in question is 0ad946c351af8b53eac06c9b8526f8e4

The key feature of CryptoShuffler is the following: instead of wasting processor time on mining, the Trojan simply substitutes the sender’s address in the clipboard! That was once the case with WebMoney and Bitcoin, but this malware sample is aimed at all popular cryptocurrencies.

As usually happens in the beginning, the Trojan writes itself into the registry for autoloading.

In later versions of the Trojan, this procedure is slightly different – if the module is implemented as a dynamically loaded library, its further run at the start is performed using the rundll32 system utility. The name of the called procedure and, concurrently, the main function of the represented library is call_directx_9.

The Trojan creates a thread of execution, in which it maintains unchanged the autorun branch specified in the screenshot above.

The substitution itself is performed using the API binding functions OpenClipboard \ GetClipboardData \ SetClipboardData

The search for the corresponding wallet in the string received from the clipboard is performed using regular expressions. Most popular cryptocurrency wallets have a fixed constant at the beginning of the string and a certain length – it is easy to create regular expressions for them. For example, the address of Bitcoin-wallets can be easily recognized by the digit “1” or “3” at the beginning of the string.

The body of the Trojan stores the wallets, corresponding to the specified cryptocurrencies. The main list looks like this.

WALLET Currency name 1v9UCfygQf3toN1vA5xyr7LhKmv9QWcwZ  BITCOIN D7uMywpgSyvy9J2RkyQ2oozT4xTmSSWGgR  DOGECOIN LeHrMiPzEUtJen73T5P1bVG2tG8PerzFR1  LITECOIN Xv4M3y36iu6Fc5ikk8XuQBDFMtRz2xFXKm  DASH 0xfb25b3d5ae0d6866da17c4de253ce439b71d0903  ETHEREUM 4ZFYNck6mZfG52RMdWThJEXq4Sjdszf719  MONERO N6VeTbNiFG1oapzPZmeLLkkNC55FQGMTgr  ??? t1VVkuasB7pNHPES2ei6LCqP1hZWb5rfPrB  ZCASH PM44dh7LNEjThgmscw8t5rb9LZqEPc2Upg  ???

The biggest profit reaches the cybercriminals’ pockets from the users of Bitcoin wallets – at the time of writing, there were ~ 23 BTC on the balance of their wallet, which at the end of October amounted to approximately $140,000. The amounts in the remaining wallets range from tens to thousands of US dollars.

The malware described is a perfect example of a “rational” gain. The scheme of its operation is simple and effective: no access to pools, no network interaction, and no suspicious processor load.



Gaza Cybergang – updated 2017 activity

Mon, 10/30/2017 - 05:00

Summary information

Gaza cybergang is an Arabic politically motivated cyber criminal group, operating since 2012 and is actively targeting the MENA (Middle East North Africa) region. Gaza cybergang attacks have never slowed down, typical targets include: governments entities/embassies, oil and gaz, media/press, activists, politicians, diplomats.

One of the interesting new facts starting from Mid-2017 is their discovery inside an Oil and Gas organization in the MENA region, infiltrating systems and pilfering data for more than a year. Another interesting finding is the usage of the recent CVE 2017-0199 vulnerability and Microsoft Access files with embedded download scripts starting, helping attackers maintain low detection rates for the latter. Traces of mobile malware are also being investigated, which started showing up from in April 2017.

Recent targets by the group does seem to be varied in nature, attackers do not seem to be selectively choosing targets, but rather seeking different kinds of MENA intelligence.

Some of the interesting new updates about Gaza cybergang:

  • Gaza cybergang attackers have continued interest in governmental entities in MENA
  • New identified targets include Oil and Gaz in MENA
  • New tools and techniques include
    • Abuse of the CVE 2017-0199 vulnerability
    • Usage of macros inside Microsoft Access files, enabling lower detection rates
    • Possible Android mobile malware being used by attackers

Previous published research:
Gaza cybergang, where’s your IR team?

Kaspersky Lab products and services successfully detect and block Gaza cybergang attacks, detection names below:

  • HEUR:Exploit.MSOffice.Generic
  • HEUR:Trojan.Win32.Cometer.gen
  • HEUR:Trojan.Win32.Generic
  • Trojan-Downloader.Win32.Downeks
  • Trojan-Spy.MSIL.Downeks
  • Win32.Bublik
  • Win32.Agentb

More information about Gaza cybergang is available to customers of Kaspersky Intelligence Reporting Service. Contact:

Technical details

Gaza cybergang attacks were previously surprisingly successful in utilising simple and common tools to achieve their goals, they rely on a variety of Remote Access Trojans (RATs), to perform their activities, including Downeks, Qasar, Cobaltstrike…

Though as recent as June 2017, attackers have started utilizing the CVE 2017-0199 vulnerability which enables direct code execution abilities from a Microsoft office document on non-patched victim systems(Cobaltstrike payload in this case). Another finding is a possible Android trojan that the attackers have positioned on one of their command servers starting from April 2017.

In most cases, malware is sent by email as compressed attachment or download links, in newer cases we have observed downloaders or Microsoft office documents with embedded macros being sent to victims starting from March 2017; when opened, the downloader would contact a URL or IP address to retrieve the actual payload. Once executed successfully, the malware grants full access to the attackers, providing them with the ability to collect files, keystrokes and screenshots from victim’s devices. If the initial downloaded malware was detected on the victim, the downloader would attempt to retrieve other malware files to victim’s device, in a attempt for one of those files to work.

The full list of indicators of compromise (IOCs) can be found in Appendix I. The list of the most interesting lure content, malware files and related droppers, command servers can be found in Appendix II.

Summary of recent campaigns

Below can be found the list of recent findings related to Gaza cybergang operations:

Command and control server Hash First seen File name/Social engineering lure upgrade.newshelpyou[.]com 552796e71f7ff304f91b39f5da46499b 25-07-2017 nvStView.exe 6fba58b9f9496cc52e78379de9f7f24e 23-03-2017 صور خاصة.exe
(Translation: Special photos) eb521caebcf03df561443194c37911a5 03-04-2017 صور خاصة.exe
(Translation: Special photos) moreoffer[.]life 66f144be4d4ef9c83bea528a4cd3baf3 27-05-2017 تصريح لأمير قطر واتهام الإمارات في اختراق وكالة الأنباء.exe
(Translation: A statement by the Emir of Qatar accusing the UAE of breaking the news agency) 3ff60c100b67697163291690e0c2c2b7 11-05-2017 MOM.InstallProxy.exe b7390bc8c8a9a71a69ce4cc0c928153b 05-04-2017 تعرف على المنقبة التي أساءت للسعودية
(Translation: Learn about the woman wearing niqab which offended Saudi) f43188accfb6923d62fe265d6d9c0940 21-03-2017 Gcc-Ksa-uae.exe 056d83c1c1b5f905d18b3c5d58ff5342 16-03-2017 مراسلة بخصوص اجتماع رؤساء البعثات.exe
(Translation: Correspondence regarding the meeting of Heads of Missions) 138.68.242[.]68 87a67371770fda4c2650564cbb00934d 20-06-2017 hamas.doc
نقاط اتفاق حماس وتيار فتح الاصلاحي.doc
(Translation: the points of agreement between Hamas and the reformist Fateh movement)
محضر اجتماع مركزية فتح الليلة.doc
(Translation: minutes of the tonight meeting)
سلفة أم راتب للموظفين يوم الثلاثاء المقبل؟.doc
(Translation: An advance on salary or full salary for employees next Tuesday?) lol.mynetav[.]org 4f3b1a2088e473c7d2373849deb4536f 20-06-2017 Notepad.exe
attachment.scr signup.updatesforme[.]club 7d3426d8eb70e4486e803afb3eeac14f 04-05-2017 Palestinian Retirement Authority Ramallah.exe 0ee4757ab9040a95e035a667457e4bc6 27-04-2017 27-4-2017 Fateh Gaza plo.exe ping.topsite[.]life b68fcf8feb35a00362758fc0f92f7c2e 19-03-2017 Downloaded by Macro in MDB files:[.]club/indexer.exe 7bef124131ffc2ef3db349b980e52847 13-03-2017 الأخ اسماعيل هنية -نائب رئيس المكتب السياسي .exe
(Translation: Brother Ismail Haniyeh – Deputy Head of the Political Bureau) d87c872869023911494305ef4acbd966 19-03-2017 Downloaded by Macro in MDB files:[.]club/wordindexer.exe a3de096598e3c9c8f3ab194edc4caa76 12-04-2017 viewimages.exe c078743eac33df15af2d9a4f24159500 28-03-2017 viewimages.exe 70d03e34cadb0f1e1bc6f4bf8486e4e8 30-03-2017 download-file.duckdns[.]org/send/Egyptian_agreement_with_President_Mahmoud_Abbas.exe 67f48fd24bae3e63b29edccc524f4096 17-04-2017 http://alasra-paper.duckdns[.]org/send/رسالة_وفد_الرئيس ابومازن_لحماس_في قطاع_غزة.rar
(Message from President Abu Mazen to Hamas in Gaza Strip) 7b536c348a21c309605fa2cd2860a41d 17-04-2017 http://alasra-paper.duckdns[.]org/send/ورقة_الاسرى_المقدمة_لفك_الاضراب .rar
(Translation: captives paper submitted to stop the strike) alasra-paper.duckdns[.]org Mobile malware N/A 23-04-2017 Possible Android malware. http://alasra-paper.duckdns[.]org/send/ hamas-wathaq.duckdns[.]org cf9d89061917e9f48481db80e674f0e9 16-04-2017 وثائق تنشر لأول مره عن حكم حماس لقطاع غزه .exe
(Translation: Documents published for the first time on Hamas ruling of Gaza Strip) manual.newphoneapp[.]com 86a89693a273d6962825cf1846c3b6ce 02-02-2017 SQLiteDatabaseBrowserPortable.exe 3f67231f30fa742138e713085e1279a6 02-02-2017 SQLiteDatabaseBrowserPortable.exe

The above listed files are further described in Appendix 1.

New findings

Gaza Cybergang attackers have been continuously evolving their skills on different levels, utilising new methods and techniques to deliver malware in addition to politically adapting social engineering decoys to regional political and humanitarian occurrences.

One of the interesting new facts starting from Mid-2017 is their discovery inside an Oil and Gas organization in the MENA region, infiltrating systems and pilfering data for more than a year, malware files found were found to be from our previously published research

While traces of Android mobile malware have been witnessed, attackers have continuously utilized the Downeks downloader and the Quasar or Cobaltstrike to target Windows devices, enabling them remote access spying and data exfiltration abilities, though now more efficient when utilizing the CVE 2017-0199 vulnerability which enables direct code execution abilities from a Microsoft office document on non-patched victim Windows systems. The use of Microsoft Access database files have also enabled the attackers to maintain low levels of detection, as it’s not an uncommon method to deliver malware.

These developments have helped the attackers continue their operations, targeting a variety of victims and organizations, sometimes even bypassing defences and persisting for prolonged periods.

1. The extended utilisation of humanitarian and political social engineering causes in the attacks

Attackers have continuously targeted victims and organizations in government entities/embassies, oil and gas, media/press, activists, politicians, diplomats.

Gaza cybergang is increasingly relying on advanced and up-to-date social engineering techniques with political and humanitarian aspects that reflect on direct regional occurrences, here is a small list of incidents that was utilized multiple time each:

  • Palestinian Government not paying salaries for Gaza employees
  • Palestinian prisoners’ hunger strike in Israeli jails
  • The political crisis in Qatar

Recent targets by the group does seem to be varied in nature, attackers do not seem to be selectively choosing targets, but rather seeking any type of intelligence.

Example lure

MD5: 66f144be4d4ef9c83bea528a4cd3baf3

تصريح لأمير قطر واتهام الإمارات في اختراق وكالة الأنباء.exe

(Translation: A statement by the Emir of Qatar accusing the UAE of breaking the news agency)

Attackers are recently utilising political events related to the Qatar political crisis in the Middle East targeting their victims.

Original filename:Qatar-27-5-2017.rar

Extracts to 66f144be4d4ef9c83bea528a4cd3baf3

تصريح لأمير قطر واتهام الإمارات في اختراق وكالة الأنباء.exe

Sha256 7fcac2f18a8844e4af9f923891cfb6f637a99195a457b6cdb916926d709c6a04

C2: moreoffer[.]life

First seen: 27 May 2017

Translation: new details on the hack of the Qatar News Agency

2. The use of Microsoft Access files with macros

Microsoft Access file with macro is another new development by the attackers group, Ms access database embedded Macros are proving to provide very low detection.

MD5: 6d6f34f7cfcb64e44d67638a2f33d619

Filename: GAZA2017.mdb


Downloads and executes:

  • data-server.cloudns[.]club/wordindexer.exe
  • data-server.cloudns[.]club/indexer.exe

Translation: database of employees not receiving salaries, click “enable content” to see data

Decrypted code

3. Exploitation of the CVE 2017-0199 vulnerability

MD5: 87a67371770fda4c2650564cbb00934d

First seen: 20-06-2017


  • hamas.doc
  • نقاط اتفاق حماس وتيار فتح الاصلاحي.doc (Translation: the points of agreement between Hamas and the reforment Fateh movement)
  • محضر اجتماع مركزية فتح الليلة.doc (Translation: minutes of the tonight Fateh meeting)
  • سلفة أم راتب للموظفين يوم الثلاثاء المقبل؟.doc (Translation: An advance on salary or full salary for employees next Tuesday?)

The attacks are typical exploitation of CVE-2017-0199 starting from an email, distributing a malicious RTF document.The vulnerability is then in the code that handles Ole2Link embedded objects, which allows Microsoft office Word to run remote files, downloaded from 138.68.242[.]68 in this case. The downloaded payload is Cobaltstrike, which then connects to lol.mynetav[.]org to receive commands from attackers. Additional details on the CVE 2017-0199 usage with Cobaltstrike by Gaza cybergang can be found here:

4. Possible Android mobile malware

Traces of APK files have been seen on one of the attackers command centers starting from 23-04-2017.

URL: http://alasra-paper.duckdns[.]org/send/%D9%88%ket-Edition-1.04_ApkHouse[.]com/Dont-Starve-Pocket-Edition-1.04_ApkHouse[.]com.apk

The file name (Dont-Starve-Pocket-Edition-1.04_ApkHouse[.]com.apk), is an Android application file hiding as a popular game. We believe the android trojan could be related to previously investigated Android trojan around Gaza strip


Gaza Cybergang has demonstrated a large number of attacks, advanced social engineering, in addition to the active development of attacks, infrastructure and the utilization of new methods and techniques. Attackers are actively improving their toolkit in an effort to minimize their exposure to security products and services. Kaspersky Lab expects these types of attacks to intensify even more both in quality and quantity in the near term.

In order to protect your company from malware, Kaspersky Lab researchers recommend implementing the following measures:

  • Educating staff to be able to distinguish spear-phishing emails or a phishing link from legitimate emails and links
  • Use proven corporate grade security solution in combination with anti-targeted attacks solutions capable of catching attacks by analyzing network anomalies
  • Providing security staff with access to latest threat intelligence data, which will arm them with helpful tools for targeted attacks prevention and discovery, such as Indicators of compromise and YARA rules
  • Making sure enterprise grade patch management processes are well established and executed.

More information about Gaza cybergang is available to customers of Kaspersky Intelligence Reporting Service. Contact:

Appendix 1: malware files descriptions and decoys

In the following, we list found description of malware files starting from March 2017, including decoys used, first dates files seen, parent files…


Parent file: 970e6188561d6c5811a8f99075888d5f

C2: moreoffer[.]life

First seen: 5 April 2017

Translation: Get to know the women wearing niqab and talking bad about the kingdom


Filename: Gcc-Ksa-uae.exe

C2: moreoffer[.]life (185.11.146[.]68)

First Seen: 21 March 2017

Translation: the permanent delegation of the cooperation council for the Arab states of the Gulf (GCC) to the United Nation and other international organizations, Geneva


مراسلة بخصوص اجتماع رؤساء البعثات.Filename: exe

Translation: Correspondence regarding the meeting of Heads of Missions (Saudi related)

Parent file: fb549e0c2fffd390ee7c4538ff30ac3e

C2: moreoffer[.]life

First Seen: 16 March 2017

Translation: The fourth foreign meeting of the Kingdom’s head of missions under the title “message of the embassador”.


Filename: 27-4-2017 Fateh Gaza plo.exe

C2: signup.updatesforme[.]club

First seen 27 April 2017

Translation: Clarification report


الأخ اسماعيل هنية -نائب رئيس المكتب السياسي .exe

(Translation: Brother Ismail Haniyah – Deputy Head of the Political Bureau)

C2: ping.topsite[.]life

First seen: 14 March 2017

Translation: Brother Ismail Haniyah – Deputy Head of the Political Bureau



C1: download-file.duckdns[.]org

C2: ping.topsite[.]life

First seen: 30 March 2017

Translation: methods to apply the palestinian national agreement pact.


C1: http://alasra-paper.duckdns[.]org/send/رسالة_وفد_الرئيس ابومازن_لحماس_في قطاع_غزة.rar

C2: ping.topsite[.]life

RAR extracts to: 5d74487ea96301a933209de3d145105d

رسالة_وفد_الرئيس ابومازن_لحماس_في قطاع_غزة.exe

First seen: 17 April 2017

Translation: a severely threatening message from Abbas’s delegation to Hamas


C1: http://alasra-paper.duckdns[.]org/send/ورقة_الاسرى_المقدمة_لفك_الاضراب .rar

Extracts to: d973135041fd26afea926e51ce141198, named (RTLO technique):

ورقة الاسرى المقدمة لفك الاضراب .exe

Translation:  captives paper submitted to stop the strike


First seen: 17 April 2017

Translation: The primary demands of the captives in the strike of freedom and dignity


وثائق تنشر لأول مره عن حكم حماس لقطاع غزه .exe     c11516cd8c797f0182d63cdf343d08ed

Translation: Documents published for the first time on Hamas ruling of Gaza Strip

C1: http://hamas-wathaq.duckdns[.]org/send/وثائق_تنشر_لأول_مره_عن_حكم_حماس_لقطاع_غزه.rar


First seen: 16 April 2017

Translation: Scandals and facts published for the first time on Hamas’s ruling of Gaza Strip

Appendix 2: List of IOCs Malicious domain names


IP addresses





Analyzing an exploit for СVE-2017-11826

Thu, 10/26/2017 - 05:00

The latest Patch Tuesday (17 October) brought patches for 62 vulnerabilities, including one that fixed СVE-2017-11826 – a critical zero-day vulnerability used to launch targeted attacks – in all versions of Microsoft Office.

The exploit for this vulnerability is an RTF document containing a DOCX document that exploits СVE-2017-11826 in the Office Open XML parser.

The exploit itself is in word/document.xml as follows:

Under the ECMA-376 standard for Office Open XML File Formats, the valid ‘font’ element describing the fonts used in the document must look like this:

In the body of the exploit the closing tag </w:font> is absent. The opening tag <w:font> is followed by the object element <o:idmap/> which cause ‘type confusion’ in the OOXML parser. Any object element can be used to successfully exploit this vulnerability. To pass one of the checks preceding the exploitation, there must be an OLEObject element in front of the <w:font> tag, and the length of the content of the attribute name must be no shorter than 32 bytes after conversion from UTF-8 into Unicode.

After conversion from UTF-8 to Unicode, E8 A3 AC E0 A2 80 becomes EC 88 88 08.

If all these conditions are fulfilled, this pointer will be dereferenced, and control will be transferred to the contents of this address with the offset 4.

To control the memory content at address 0x088888EC, the attackers apply the popular heap spraying technique with use of ActiveX components:

The exploit bypasses ASLR and DEP using ROP and gadgets from msvbvm60.dll. The msvbvm60.dll module is loaded from the RTF document with the help of a CLSID associated with this DLL:

The first part of ROP sets the ESP register’s value:

The second part of ROP is ignored: it was used to set the EIP register at 0x088883EC. The last ‘pop eax; retn’ gadget moves the address 0x729410D0 into EAX. This is the address for the VirtualProtect pointer in the Imports area of msvbvm60.dll from Kernel32.dll:

The VirtualProtect pointer is used in the next ROP gadget to call the function VirtualProtect(0x8888C90, 0x201, 0x40, 0x72A4C045). After this, control is transferred to the shellcode at the address 0x8888F70, which decrypts and executes the embedded:

Kaspersky Lab’s security solutions detect exploits for СVE-2017-11826 as:

  • MSWord.Agent.ix;
  • MSOffice.CVE-2017-11826.a;
  • HEUR:Exploit.MSOffice.Generic.



Bad Rabbit ransomware

Tue, 10/24/2017 - 14:16

What happened?

On October 24th we observed notifications of mass attacks with ransomware called Bad Rabbit. It has been targeting organizations and consumers, mostly in Russia but there have also been reports of victims in Ukraine. Here’s what a ransom message looks like for the unlucky victims:

What is bad rabbit?

Bad Rabbit is a previously unknown ransomware family.

How is bad rabbit distributed?

The ransomware dropper was distributed with the help of drive-by attacks. While the target is visiting a legitimate website, a malware dropper is being downloaded from the threat actor’s infrastructure. No exploits were used, so the victim would have to manually execute the malware dropper, which pretends to be an Adobe Flash installer.

We’ve detected a number of compromised websites, all of which were news or media websites.

Whom does it target?

Most of the targets are located in Russia. Similar but fewer attacks have also been seen in other countries – Ukraine, Turkey and Germany. Overall, there are almost 200 targets, according to the KSN statistics.

Since when does Kaspersky Lab detect the threat?

We have been proactively detecting the original vector attack since it began on the morning of October 24. The attack lasted until midday, although we are still detecting ongoing attacks.

How is it different to ExPetr? Or it is the same malware?

Our observations suggest that this been a targeted attack against corporate networks, using methods similar to those used during the ExPetr attack.

Technical details

According to our telemetry, the ransomware is spread via a drive-by attack.

The ransomware dropper is distributed from hxxp://1dnscontrol[.]com/flash_install.php

Also according to our telemetry data, victims are redirected to this malware web resource from legitimate news websites.

The downloaded file named install_flash_player.exe needs to be manually launched by the victim. To operate correctly, it needs elevated administrative privileges which it attempts to obtain using the standard UAC prompt. If started, it will save the malicious DLL as C:\Windows\infpub.dat and launch it using rundll32.

Pseudocode of the procedure that installs the malicious DLL

infpub.dat appears to be capable of brute-forcing NTLM login credentials to Windows machines that have pseudo-random IP addresses.

The hard-coded list of credentials

infpub.dat will also install the malicious executable dispci.exe into C:\Windows and create a task to launch it.

Pseudocode of the procedure that creates the task which launches the malicious executable

What’s more, infpub.dat acts as a typical file encrypting ransomware: it finds the victim’s data files using an embedded extension list and encrypts them using the criminal’s public RSA-2048 key.

The public key of the criminals and the extension list

The criminal’s public key parameters:

Public-Key: (2048 bit)
Exponent: 65537 (0x10001)

The executable dispci.exe appears to be derived from the code base of the legitimate utility DiskCryptor. It acts as the disk encryption module which also installs the modified bootloader and prevents the normal boot-up process of the infected machine.

An interesting detail that we noticed when analyzing the sample of this threat: it looks like the criminals behind this malware are fans of the famous books & TV show series Game Of Thrones. Some of the strings used throughout the code are the names of different characters from this series.

Dragon names from Game Of Thrones

Character name from Game Of Thrones

Kaspersky Lab experts are working on a detailed analysis of this ransomware to find possible flaws in its cryptographic routines.

Kaspersky Lab corporate customers are also advised to:

  • make sure that all protection mechanisms are activated as recommended; and that KSN and System Watcher components (which are enabled by default) are not disabled.
  • update the antivirus databases immediately.

The abovementioned measures should be sufficient. However, as additional precautions we advise the following:

  • restricting execution of files with the paths c:\windows\infpub.dat and C:\Windows\cscc.dat in Kaspersky Endpoint Security.
  • configuring and enabling Default Deny mode in the Application Startup Control component of Kaspersky Endpoint Security to ensure and enforce proactive defense against this and other attacks.

Kaspersky Lab products detect this threat with the following verdicts:

  • Trojan-Ransom.Win32.Gen.ftl
  • Multi.Generic
  • PDM:Trojan.Win32.Generic

fbbdc39af1139aebba4da004475e8839 – install_flash_player.exe
1d724f95c61f1055f0d02c2154bbccd3 – C:\Windows\infpub.dat
b14d8faf7f0cbcfad051cefe5f39645f – C:\Windows\dispci.exe


Dangerous liaisons

Tue, 10/24/2017 - 05:00

It seems just about everyone has written about the dangers of online dating, from psychology magazines to crime chronicles. But there is one less obvious threat not related to hooking up with strangers – and that is the mobile apps used to facilitate the process. We’re talking here about intercepting and stealing personal information and the de-anonymization of a dating service that could cause victims no end of troubles – from messages being sent out in their names to blackmail. We took the most popular apps and analyzed what sort of user data they were capable of handing over to criminals and under what conditions.

We studied the following online dating applications:

By de-anonymization we mean the user’s real name being established from a social media network profile where use of an alias is meaningless.

User tracking capabilities

First of all, we checked how easy it was to track users with the data available in the app. If the app included an option to show your place of work, it was fairly easy to match the name of a user and their page on a social network. This in turn could allow criminals to gather much more data about the victim, track their movements, identify their circle of friends and acquaintances. This data can then be used to stalk the victim.

Discovering a user’s profile on a social network also means other app restrictions, such as the ban on writing each other messages, can be circumvented. Some apps only allow users with premium (paid) accounts to send messages, while others prevent men from starting a conversation. These restrictions don’t usually apply on social media, and anyone can write to whomever they like.

More specifically, in Tinder, Happn and Bumble users can add information about their job and education. Using that information, we managed in 60% of cases to identify users’ pages on various social media, including Facebook and LinkedIn, as well as their full names and surnames.

An example of an account that gives workplace information that was used to identify the user on other social media networks

In Happn for Android there is an additional search option: among the data about the users being viewed that the server sends to the application, there is the parameter fb_id – a specially generated identification number for the Facebook account. The app uses it to find out how many friends the user has in common on Facebook. This is done using the authentication token the app receives from Facebook. By modifying this request slightly – removing some of the original request and leaving the token – you can find out the name of the user in the Facebook account for any Happn users viewed.

Data received by the Android version of Happn

It’s even easier to find a user account with the iOS version: the server returns the user’s real Facebook user ID to the application.

Data received by the iOS version of Happn

Information about users in all the other apps is usually limited to just photos, age, first name or nickname. We couldn’t find any accounts for people on other social networks using just this information. Even a search of Google images didn’t help. In one case the search recognized Adam Sandler in a photo, despite it being of a woman that looked nothing like the actor.

The Paktor app allows you to find out email addresses, and not just of those users that are viewed. All you need to do is intercept the traffic, which is easy enough to do on your own device. As a result, an attacker can end up with the email addresses not only of those users whose profiles they viewed but also for other users – the app receives a list of users from the server with data that includes email addresses. This problem is found in both the Android and iOS versions of the app. We have reported it to the developers.

Fragment of data that includes a user’s email address

Some of the apps in our study allow you to attach an Instagram account to your profile. The information extracted from it also helped us establish real names: many people on Instagram use their real name, while others include it in the account name. Using this information, you can then find a Facebook or LinkedIn account.


Most of the apps in our research are vulnerable when it comes to identifying user locations prior to an attack, although this threat has already been mentioned in several studies (for instance, here and here). We found that users of Tinder, Mamba, Zoosk, Happn, WeChat, and Paktor are particularly susceptible to this.

Screenshot of the Android version of WeChat showing the distance to users

The attack is based on a function that displays the distance to other users, usually to those whose profile is currently being viewed. Even though the application doesn’t show in which direction, the location can be learned by moving around the victim and recording data about the distance to them. This method is quite laborious, though the services themselves simplify the task: an attacker can remain in one place, while feeding fake coordinates to a service, each time receiving data about the distance to the profile owner.

Mamba for Android displays the distance to a user

Different apps show the distance to a user with varying accuracy: from a few dozen meters up to a kilometer. The less accurate an app is, the more measurements you need to make.

As well as the distance to a user, Happn shows how many times “you’ve crossed paths” with them

Unprotected transmission of traffic

During our research, we also checked what sort of data the apps exchange with their servers. We were interested in what could be intercepted if, for example, the user connects to an unprotected wireless network – to carry out an attack it’s sufficient for a cybercriminal to be on the same network. Even if the Wi-Fi traffic is encrypted, it can still be intercepted on an access point if it’s controlled by a cybercriminal.

Most of the applications use SSL when communicating with a server, but some things remain unencrypted. For example, Tinder, Paktor and Bumble for Android and the iOS version of Badoo upload photos via HTTP, i.e., in unencrypted format. This allows an attacker, for example, to see which accounts the victim is currently viewing.

HTTP requests for photos from the Tinder app

The Android version of Paktor uses the quantumgraph analytics module that transmits a lot of information in unencrypted format, including the user’s name, date of birth and GPS coordinates. In addition, the module sends the server information about which app functions the victim is currently using. It should be noted that in the iOS version of Paktor all traffic is encrypted.

The unencrypted data the quantumgraph module transmits to the server includes the user’s coordinates

Although Badoo uses encryption, its Android version uploads data (GPS coordinates, device and mobile operator information, etc.) to the server in an unencrypted format if it can’t connect to the server via HTTPS.

Badoo transmitting the user’s coordinates in an unencrypted format

The Mamba dating service stands apart from all the other apps. First of all, the Android version of Mamba includes a flurry analytics module that uploads information about the device (producer, model, etc.) to the server in an unencrypted format. Secondly, the iOS version of the Mamba application connects to the server using the HTTP protocol, without any encryption at all.

Mamba transmits data in an unencrypted format, including messages

This makes it easy for an attacker to view and even modify all the data that the app exchanges with the servers, including personal information. Moreover, by using part of the intercepted data, it is possible to gain access to account management.

Using intercepted data, it’s possible to access account management and, for example, send messages

Mamba: messages sent following the interception of data

Despite data being encrypted by default in the Android version of Mamba, the application sometimes connects to the server via unencrypted HTTP. By intercepting the data used for these connections, an attacker can also get control of someone else’s account. We reported our findings to the developers, and they promised to fix these problems.

An unencrypted request by Mamba

We also managed to detect this in Zoosk for both platforms – some of the communication between the app and the server is via HTTP, and the data is transmitted in requests, which can be intercepted to give an attacker the temporary ability to manage the account. It should be noted that the data can only be intercepted at that moment when the user is loading new photos or videos to the application, i.e., not always. We told the developers about this problem, and they fixed it.

Unencrypted request by Zoosk

In addition, the Android version of Zoosk uses the mobup advertising module. By intercepting this module’s requests, you can find out the GPS coordinates of the user, their age, sex, model of smartphone – all this is transmitted in unencrypted format. If an attacker controls a Wi-Fi access point, they can change the ads shown in the app to any they like, including malicious ads.

An unencrypted request from the mopub ad unit also contains the user’s coordinates

The iOS version of the WeChat app connects to the server via HTTP, but all data transmitted in this way remains encrypted.

Data in SSL

In general, the apps in our investigation and their additional modules use the HTTPS protocol (HTTP Secure) to communicate with their servers. The security of HTTPS is based on the server having a certificate, the reliability of which can be verified. In other words, the protocol makes it possible to protect against man-in-the-middle attacks (MITM): the certificate must be checked to ensure it really does belong to the specified server.

We checked how good the dating apps are at withstanding this type of attack. This involved installing a ‘homemade’ certificate on the test device that allowed us to ‘spy on’ the encrypted traffic between the server and the application, and whether the latter verifies the validity of the certificate.

It’s worth noting that installing a third-party certificate on an Android device is very easy, and the user can be tricked into doing it. All you need to do is lure the victim to a site containing the certificate (if the attacker controls the network, this can be any resource) and convince them to click a download button. After that, the system itself will start installation of the certificate, requesting the PIN once (if it is installed) and suggesting a certificate name.

Everything’s a lot more complicated with iOS. First, you need to install a configuration profile, and the user needs to confirm this action several times and enter the password or PIN number of the device several times. Then you need to go into the settings and add the certificate from the installed profile to the list of trusted certificates.

It turned out that most of the apps in our investigation are to some extent vulnerable to an MITM attack. Only Badoo and Bumble, plus the Android version of Zoosk, use the right approach and check the server certificate.

It should be noted that though WeChat continued to work with a fake certificate, it encrypted all the transmitted data that we intercepted, which can be considered a success since the gathered information can’t be used.

Message from Happn in intercepted traffic

Remember that most of the programs in our study use authorization via Facebook. This means the user’s password is protected, though a token that allows temporary authorization in the app can be stolen.

Token in a Tinder app request

A token is a key used for authorization that is issued by the authentication service (in our example Facebook) at the request of the user. It is issued for a limited time, usually two to three weeks, after which the app must request access again. Using the token, the program gets all the necessary data for authentication and can authenticate the user on its servers by simply verifying the credibility of the token.

Example of authorization via Facebook

It’s interesting that Mamba sends a generated password to the email address after registration using the Facebook account. The same password is then used for authorization on the server. Thus, in the app, you can intercept a token or even a login and password pairing, meaning an attacker can log in to the app.

App files (Android)

We decided to check what sort of app data is stored on the device. Although the data is protected by the system, and other applications don’t have access to it, it can be obtained with superuser rights (root). Because there are no widespread malicious programs for iOS that can get superuser rights, we believe that for Apple device owners this threat is not relevant. So only Android applications were considered in this part of the study.

Superuser rights are not that rare when it comes to Android devices. According to KSN, in the second quarter of 2017 they were installed on smartphones by more than 5% of users. In addition, some Trojans can gain root access themselves, taking advantage of vulnerabilities in the operating system. Studies on the availability of personal information in mobile apps were carried out a couple of years ago and, as we can see, little has changed since then.

Analysis showed that most dating applications are not ready for such attacks; by taking advantage of superuser rights, we managed to get authorization tokens (mainly from Facebook) from almost all the apps. Authorization via Facebook, when the user doesn’t need to come up with new logins and passwords, is a good strategy that increases the security of the account, but only if the Facebook account is protected with a strong password. However, the application token itself is often not stored securely enough.

Tinder app file with a token

Using the generated Facebook token, you can get temporary authorization in the dating application, gaining full access to the account. In the case of Mamba, we even managed to get a password and login – they can be easily decrypted using a key stored in the app itself.

Mamba app file with encrypted password

Most of the apps in our study (Tinder, Bumble, OK Cupid, Badoo, Happn and Paktor) store the message history in the same folder as the token. As a result, once the attacker has obtained superuser rights, they will have access to correspondence.

Paktor app database with messages

In addition, almost all the apps store photos of other users in the smartphone’s memory. This is because apps use standard methods to open web pages: the system caches photos that can be opened. With access to the cache folder, you can find out which profiles the user has viewed.


Having gathered together all the vulnerabilities found in the studied dating apps, we get the following table:

App Location Stalking HTTP (Android) HTTP (iOS) HTTPS Messages Token Tinder + 60% Low Low + + + Bumble – 50% Low NO – + + OK Cupid – 0% NO NO + + + Badoo – 0% Medium NO – + + Mamba + 0% High High + – + Zoosk + 0% High High –
(+ iOS) – + Happn + 100% NO NO + + + WeChat + 0% NO NO – – – Paktor + 100% emails Medium NO + + +

Location — determining user location (“+” – possible, “-” not possible)

Stalking — finding the full name of the user, as well as their accounts in other social networks, the percentage of detected users (percentage indicates the number of successful identifications)

HTTP — the ability to intercept any data from the application sent in an unencrypted form (“NO” – could not find the data, “Low” – non-dangerous data, “Medium” – data that can be dangerous, “High” – intercepted data that can be used to get account management).

HTTPS — interception of data transmitted inside the encrypted connection (“+” – possible, “-” not possible).

Messages — access to user messages by using root rights (“+” – possible, “-” not possible).

TOKEN — possibility to steal authentication token by using root rights (“+” – possible, “-” not possible).

As you can see from the table, some apps practically do not protect users’ personal information. However, overall, things could be worse, even with the proviso that in practice we didn’t study too closely the possibility of locating specific users of the services. Of course, we are not going to discourage people from using dating apps, but we would like to give some recommendations on how to use them more safely. First, our universal advice is to avoid public Wi-Fi access points, especially those that are not protected by a password, use a VPN, and install a security solution on your smartphone that can detect malware. These are all very relevant for the situation in question and help prevent the theft of personal information. Secondly, do not specify your place of work, or any other information that could identify you. Safe dating!

ATM malware is being sold on Darknet market

Tue, 10/17/2017 - 05:00

Disclaimer and warning

ATM systems appear to be very secure, but the money can be accessed fairly easily if you know what you are doing. Criminals are exploiting hardware and software vulnerabilities to interact with ATMs, meaning they need to be made more secure. This can be achieved with the help of additional security software, properly configured to stop the execution of non-whitelisted programs on ATMs.

Worryingly, it is very easy to find detailed manuals of ATM malware. Anybody can simply buy them for around 5000 USD on darknet markets.


In May 2017, Kaspersky Lab researchers discovered a forum post advertising ATM malware that was targeting specific vendor ATMs. The forum contained a short description of a crimeware kit designed to empty ATMs with the help of a vendor specific API, without interacting with ATM users and their data. The post links to an offer that was initially published on the AlphaBay Darknet marketplace, which was recently taken down by the FBI.

Advertisement post

An offer post on AlphaBay market

The price of the kit was 5000 USD at the time of research. The AlphaBay description includes details such as the required equipment, targeted ATMs models, as well as tips and tricks for the malware’s operation. And part of a detailed manual for the toolkit was also provided.

Screenshot of a description on AlphaBay market

Previously described ATM malware Tyupkin was also mentioned in this text. The manual “Wall ATM Read Me.txt” was distributed as a plain text file, written in poor English and with bad text formatting. The use of slang and grammatical mistakes suggests that this text was most likely written by a native Russian-speaker.

Apart of a manual with text formatting applied

The manual provides a detailed picture, though only a fragment of the complete manual is being shown. There is a description for each step of the dispense process:

Prepare an all tools, all the programs should be placed on a flash disk.
Tools are wireless keyboard, usb hub, usb cable, usb adapter usb a female to b female, Windows 7 laptop or a tablet ( to run code generator) and a drill.
Find an appropriate ATM
Open ATM door and plug into USB port.
Execute Stimulator to see full information of all the ATM cassettes.
Execute CUTLET MAKER to get it is code.
Execute password generator on a tablet or on a laptop and paste CUTLET MAKER code to it, put the result password to CUTLET MAKER.
Dispense the money from chosen cassette.

The manual provides usage descriptions for all parts of the toolset. The list of crimeware from the kit consists of CUTLET MAKER ATM malware, the primary element, with a password generator included and a Stimulator – an application to gather cash cassette statuses of a target ATM. The crimeware kit is a collection of programs possibly written by different authors, though CUTLET MAKER and Stimulator were protected in the same way, c0decalc is a simple terminal-based application without any protection at all.

Delicious cutlet ingredients: CUTLET MAKER, c0decalc and Stimulator

The first sample was named “CUTLET MAKER” by its authors and has been designed to operate the cash dispense process on specific vendor ATMs.

To answer the question of how a cook from the CUTLET MAKER interface and cutlets relate to stealing money from ATMs, we must explain the meaning of the word “Cutlet“. Originally, it means a meat dish, but as a Russian slang term “Cutlet” (котлета) means “a bundle of money”, suggesting that the criminals behind the malware might be native Russian speakers.

The “Cutlet Maker” malware functionality suggests that two people are supposed to be involved in the theft – the roles are called “drop” and “drop master”. Access to the dispense mechanism of CUTLET MAKER is password protected. Though there could be just one person with the c0decalc application needed to generate a password. Either network or physical access to an ATM is required to enter the code in the application text area and also to interact with the user interface.

Stimulator was possibly developed by the same authors. Its purpose is to retrieve and show the status information of specific vendor ATM cash cassettes (such as currency, value and the amount of notes).

CUTLET MAKER and c0decalc

CUTLET MAKER is the main module responsible for dispensing money from the ATM. The sample analysed in this research has the MD5 checksum “fac356509a156a8f11ce69f149198108” and the compilation timestamp Sat Jul 30 20:17:08 2016 UTC.

The program is written in Delphi and was packed with VMProtect, however it is possible that multiple packers might have been used.

Different versions of the main component were found while researching this toolset. The first known submission of the first version sent to a public multiscanner service took place on June 22nd 2016. All submissions discovered by Kaspersky Lab were performed from different countries, with Ukraine being the chronological first country of origin.

Known CUTLET MAKER filenames (according to public multiscanner service information):


The following version information was captured from the application’s window caption, followed after a “CUTLET MAKER” name. Known versions at the time of research were:

1.0 F

The assumed development period is from 2016-06-22 to 2016-08-18, according to the first submission date of the earliest version and the last submission date of the latest version at the time of writing. The application requires a special library to operate, which is part of a proprietary ATM API, controlling the cash dispenser unit.

With all the dependencies in place, the interface shows a code.

CUTLET MAKER challenge code marked with red rectangle

In order to unlock the application, a password from c0decalc generator needs to be entered, thereby answering the given challenge code. If the password is incorrect, the interface won’t react to any further input.

Each “CHECK HEAT” and “start cooking!” button corresponds to a specific ATM cash cassette. Buttons labeled “CHECK HEAT” dispense one note, “start cooking!” dispenses 50 “cutlets” with 60 notes each.  The “Stop!” button stops an ongoing “start cooking!” process. “Reset” is intended to reset the dispense process.

c0decalc a password generator for CUTLET MAKER

This tool is an unprotected command line application, written in Visual C. The purpose of this application is to generate a password for CUTLET MAKER’s graphical interface.

The compilation timestamp for this specific sample is Sun Nov 13 11:35:25 2016 UTC and was first uploaded to a public multiscanner service on December 7th 2016.

Example output for “12345678” input

Kaspersky Lab researchers checked the algorithm during the analysis and found “CUTLET MAKER” working with the passwords generated by “c0decalc”.


The Stimulator sample analysed in this research has the MD5 hash “27640bb7908ca7303d13d50c14ccf669”. This sample is also written in Delphi and packed the same way as “CUTLET MAKER”. The compilation timestamp is Sat Jul 16 18:34:47 2016 UTC.

The application is designed to work on specific vendor ATMs and also uses proprietary API calls.

Some additional symbols were found in the memory dump of a “Stimulator” process, pointing to an interesting part of the application. After execution and pressing the “STIMULATE ME!” button, the proprietary API function is used to fetch an ATM’s cassette status. The following cassette state results are used:


Each preceding number is mapped to an ATM cassette. The three character states are interpreted as follows:

nCUR cassette n currency (like “USD”, “RUB”) nVAL cassette n note value (like 00000005, 00000020 ) nACT cassette n counter for specific notes in a cassette (value from 0 to 3000) nNDV number of notes in the ATM for cassette n (value from 0 to 3000)

The result of “STIMULATE ME!” button press in proper environment

Each column, shown in the picture above, describes the state of one corresponding ATM cassette.

The background picture used in the application interface turns out to be quite unique, the original photo was posted on a DIY blog:

Original picture as used in “Stimulator” application (photo by Alex Santantonio)


This type of malware does not affect bank customers directly, it is intended for the theft of cash from specific vendor ATMs. CUTLET MAKER and Stimulator show how criminals are using legitimate proprietary libraries and a small piece of code to dispense money from an ATM. Examples of appropriate countermeasures against such attacks include default-deny policies and device control. The first measure prevents criminals from running their own code on the ATM’s internal PC. It is likely that ATMs in these attacks were infected through physical access to the PC, which means criminals were using USB drives to install malware onto the machine. In such a case, device control software would prevent them from connecting new devices, such as USB sticks. Kaspersky Embedded Systems Security will help to extend the security level of ATMs.

Kaspersky Lab products detects this treats as Backdoor.Win32.ATMletcut, Backdoor.Win32.ATMulator, Trojan.Win32.Agent.ikmo

BlackOasis APT and new targeted attacks leveraging zero-day exploit

Mon, 10/16/2017 - 10:28

More information about BlackOasis APT is available to customers of Kaspersky Intelligence Reporting Service. Contact:


Kaspersky Lab has always worked closely with vendors to protect users. As soon as we find new vulnerabilities we immediately inform the vendor in a responsible manner and provide all the details required for a fix.

On October 10, 2017, Kaspersky Lab’s advanced exploit prevention systems identified a new Adobe Flash zero day exploit used in the wild against our customers. The exploit was delivered through a Microsoft Office document and the final payload was the latest version of FinSpy malware. We have reported the bug to Adobe who assigned it CVE-2017-11292 and released a patch earlier today:

So far only one attack has been observed in our customer base, leading us to believe the number of attacks are minimal and highly targeted.

Analysis of the payload allowed us to confidently link this attack to an actor we track as “BlackOasis”. We are also highly confident that BlackOasis was also responsible for another zero day exploit (CVE-2017-8759) discovered by FireEye in September 2017.  The FinSpy payload used in the current attacks (CVE-2017-11292) shares the same command and control (C2) server as the payload used with CVE-2017-8759 uncovered by FireEye.

BlackOasis Background

We first became aware of BlackOasis’ activities in May 2016, while investigating another Adobe Flash zero day. On May 10, 2016, Adobe warned of a vulnerability (CVE-2016-4117) affecting Flash Player and earlier versions for Windows, Macintosh, Linux, and Chrome OS. The vulnerability was actively being exploited in the wild.

Kaspersky Lab was able to identify a sample exploiting this vulnerability that was uploaded to a multi scanner system on May 8, 2016. The sample, in the form of an RTF document, exploited CVE-2016-4117 to download and install a program from a remote C&C server. Although the exact payload of the attack was no longer in the C&C, the same server was hosting multiple FinSpy installation packages.

Leveraging data from Kaspersky Security Network, we identified two other similar exploit chains used by BlackOasis in June 2015 which were zero days at the time.  Those include CVE-2015-5119 and CVE-2016-0984, which were patched in July 2015 and February 2016 respectively.  These exploit chains also delivered FinSpy installation packages.

Since the discovery of BlackOasis’ exploitation network, we’ve been tracking this threat actor with the purpose of better understanding their operations and targeting and have seen a couple dozen new attacks. Some lure documents used in these attacks are shown below:

Decoy documents used in BlackOasis attacks

To summarize, we have seen BlackOasis utilizing at least five zero days since June 2015:</p style=”margin-bottom:0!important”>

  • CVE-2015-5119 – June 2015
  • CVE-2016-0984 – June 2015
  • CVE-2016-4117 – May 2016
  • CVE-2017-8759 – Sept 2017
  • CVE-2017-11292 – Oct 2017
Attacks Leveraging CVE-2017-11292

The attack begins with the delivery of an Office document, presumably in this instance via e-mail.  Embedded within the document is an ActiveX object which contains the Flash exploit.

Flash object in the .docx file, stored in uncompressed format

The Flash object contains an ActionScript which is responsible for extracting the exploit using a custom packer seen in other FinSpy exploits.

Unpacking routine for SWF exploit

The exploit is a memory corruption vulnerability that exists in the “com.adobe.tvsdk.mediacore.BufferControlParameters” class.  If the exploit is successful, it will gain arbitrary read / write operations within memory, thus allowing it to execute a second stage shellcode.

The first stage shellcode contains an interesting NOP sled with alternative instructions, which was most likely designed in such a way to avoid detection by antivirus products looking for large NOP blocks inside flash files:

NOP sled composed of 0x90 and 0x91 opcodes

The main purpose of the initial shellcode is to download second stage shellcode from hxxp://89.45.67[.]107/rss/5uzosoff0u.iaf.

Second stage shellcode

The second stage shellcode will then perform the following actions:</p style=”margin-bottom:0!important”>

  1. Download the final payload (FinSpy) from hxxp://89.45.67[.]107/rss/mo.exe
  2. Download a lure document to display to the victim from the same IP
  3. Execute the payload and display the lure document
Payload – mo.exe

As mentioned earlier, the “mo.exe” payload (MD5: 4a49135d2ecc07085a8b7c5925a36c0a) is the newest version of Gamma International’s FinSpy malware, typically sold to nation states and other law enforcement agencies to use in lawful surveillance operations.  This newer variant has made it especially difficult for researchers to analyze the malware due to many added anti-analysis techniques, to include a custom packer and virtual machine to execute code.

The PCODE of the virtual machine is packed with the aplib packer.

Part of packed VM PCODE

After unpacking, the PCODE it will look like the following:

Unpacked PCODE

After unpacking the virtual machine PCODE is then decrypted:

Decrypted VM PCODE

The custom virtual machine supports a total of 34 instructions:

Example of parsed PCODE

In this example, the “1b” instruction is responsible for executing native code that is specified in parameter field.

Once the payload is successfully executed, it will proceed to copy files to the following locations:</p style=”margin-bottom:0!important”>

  • C:\ProgramData\ManagerApp\AdapterTroubleshooter.exe
  • C:\ProgramData\ManagerApp\
  • C:\ProgramData\ManagerApp\
  • C:\ProgramData\ManagerApp\msvcr90.dll
  • C:\ProgramData\ManagerApp\d3d9.dll

The “AdapterTroubleshooter.exe” file is a legitimate binary which is leveraged to use the famous DLL search order hijacking technique.  The “d3d9.dll” file is malicious and is loaded into memory by the legit binary upon execution.  Once loaded, the DLL will then inject FinSpy into the Winlogon process.

Part of injected code in winlogon process

The payload calls out to three C2 servers for further control and exfiltration of data. We have observed two of them used in the past with other FinSpy payloads. Most recently one of these C2 servers was used together with CVE-2017-8759 in the attacks reported by FireEye in September 2017. These IPs and other previous samples tie closely to the BlackOasis APT cluster of FinSpy activity.

Targeting and Victims

BlackOasis’ interests span a wide gamut of figures involved in Middle Eastern politics and verticals disproportionately relevant to the region. This includes prominent figures in the United Nations, opposition bloggers and activists, and regional news correspondents. During 2016, we observed a heavy interest in Angola, exemplified by lure documents indicating targets with suspected ties to oil, money laundering, and other illicit activities. There is also an interest in international activists and think tanks.

Victims of BlackOasis have been observed in the following countries: Russia, Iraq, Afghanistan, Nigeria, Libya, Jordan, Tunisia, Saudi Arabia, Iran, Netherlands, Bahrain, United Kingdom and Angola.


We estimate that the attack on HackingTeam in mid-2015 left a gap on the market for surveillance tools, which is now being filled by other companies. One of these is Gamma International with their FinFisher suite of tools. Although Gamma International itself was hacked by Phineas Fisher in 2014, the breach was not as serious as it was in the case of HackingTeam. Additionally, Gamma had two years to recover from the attack and pick up the pace.

We believe the number of attacks relying on FinFisher software, supported by zero day exploits such as the ones described here will continue to grow.

What does it mean for everyone and how to defend against such attacks, including zero-day exploits?

For CVE-2017-11292 and other similar vulnerabilities, one can use the killbit for Flash within their organizations to disable it in any applications that respect it.  Unfortunately, doing this system-wide is not easily done, as Flash objects can be loaded in applications that potentially do not follow the killbit. Additionally, this may break any other necessary resources that rely on Flash and of course, it will not protect against exploits for other third party software.

Deploying a multi-layered approach including access policies, anti-virus, network monitoring and whitelisting can help ensure customers are protected against threats such as this.  Users of Kaspersky products are protected as well against this threat by one of the following detections:</p style=”margin-bottom:0!important”>

  • PDM:Exploit.Win32.Generic
  • HEUR:Exploit.SWF.Generic
  • HEUR:Exploit.MSOffice.Generic

More information about BlackOasis APT is available to customers of Kaspersky Intelligence Reporting Service. Contact:


We would like to thank the Adobe Product Security Incident Response Team (PSIRT) for working with us to identify and patch this vulnerability.

  1. Adobe Bulletin
Indicators of compromise


ATMii: a small but effective ATM robber

Tue, 10/10/2017 - 05:00

While some criminals blow up ATMs to steal cash, others use less destructive methods, such as infecting the ATM with malware and then stealing the money. We have written about this phenomenon extensively in the past and today we can add another family of malware to the list – Backdoor.Win32.ATMii.

ATMii was first brought to our attention in April 2017, when a partner from the financial industry shared some samples with us. The malware turned out to be fairly straightforward, consisting of only two modules: an injector module (exe.exe, 3fddbf20b41e335b6b1615536b8e1292) and the module to be injected (dll.dll, dc42ed8e1de55185c9240f33863a6aa4). To use this malware, criminals need direct access to the target ATM, either over the network or physically (e.g. over USB). ATMii, if it is successful, allows criminals to dispense all the cash from the ATM.

exe.exe – an injector and control module

The injector is an unprotected command line application, written in Visual C with a compilation timestamp: Fri Nov 01 14:33:23 2013 UTC. Since this compilation timestamp is from 4 years ago – and we do not think this threat could have gone unnoticed for 4 years – we believe it is a fake timestamp. What’s also interesting is the OS that is supported by the malware: One more recent than Windows XP. We can see this in the image below, where the first argument for the OpenProcess() function is 0x1FFFFu.

OpenProcess call with the PROCESS_ALL_ACCESS constant

It is the PROCESS_ALL_ACCESS constant, but this constant value differs in older Windows versions such as Windows XP (see the picture below). This is interesting because most ATMs still run on Windows XP, which is thus not supported by the malware.

A list of PROCESS_ALL_ACCESS values per Windows version

The injector, which targets the atmapp.exe (proprietary ATM software) process, is fairly poorly written, since it depends on several parameters. If none are given, the application catches an exception. The parameters are pretty self-explanatory:

param  short description /load Tries to inject dll.dll into atmapp.exe process /cmd Creates/Updates C:\ATM\c.ini file to pass commands and params to infected library /unload Tries to unload injected library from atmapp.exe process, while restoring its state.

/load param

<exe.exe> /load

The application searches for a process with the name atmapp.exe and injects code into it that loads the “dll.dll” library (which has to be in the same folder as the exe.exe file). After it has been loaded it calls the DLLmain function.

/unload param <exe.exe> /unload

As the name already suggests, it is the opposite of the /load parameter; it unloads the injected module and restores the process to its original state.

/cmd param <exe.exe> /cmd [cmd] [params]

The application creates/updates C:\ATM\c.ini which is used by the injected DLL to read commands. The file is updated each time the .exe is run with the /cmd param.

Contents of c.ini after execution of “exe.exe /cmd info”

The executable understands the following set of commands:

command description scan Scans for the CASH_UNIT XFS service disp Stands for “dispense”. The injected module should dispense “amount” cash of “currency” (amount and currency are used as parameters) info Gets info about ATM cash cassettes, all the returned data goes to the log file. die Injected module removes C:\ATM\c.ini file dll.dll injecting module

After injection and execution of the DllMain function, the dll.dll library loads msxfs.dll and replaces the WFSGetInfo function with a special wrap function, named mWFSGetInfo.

At the time of the first call to the fake WFSGetInfo function, C:\ATM\c.ini is ignored and the library tries to find the ATM’s CASH_UNIT service id and stores the result, basically in the same way as the scan command does. If the CASH_UNIT service is not found, dll.dll won’t function. However, if successful, all further calls go to the mWFSGetInfo function, which performs the additional logic (reading, parsing and executing the commands from the C:\ATM\c.ini file).

Contents of C:\ATM\c.ini after execution of “exe.exe /cmd disp RUB 6000”

Below is an output of the strings program uncovering some interesting log messages and the function names to be imported. The proprietary MSXFS.DLL library and its functions used in the ATMii malware are marked with red boxes.

“scan” command

Because of the architecture of XFS, which is divided into services, the injected library first needs to find the dispense service. This command must be successfully called, because the disp and info commands depend on the service id retrieved by scan. Scan is automatically called after the dll has been injected into atmapp.exe.

After collecting the WFS_INF_CDM_STATUS data, additional data gets added to the tlogs.log. An example can be found below:

(387):cmd_scan() Searching valid service
(358):FindValidService() Checking device index=0
(70):CheckServiceForValid() ————————————————
(72):CheckServiceForValid() Waiting for lock
(76):CheckServiceForValid() Device was locked
(86):CheckServiceForValid() WFSGetInfo Success 0
(182):CheckServiceForValid() Done-> szDevice: WFS_CDM_DEVONLINE, szDispenser: WFS_CDM_DISPOK, szIntermediateStacker: WFS_CDM_ISEMPTY, szSafeDoor: WFS_CDM_DOORCLOSED
(195):CheckServiceForValid() Unlocking device
(390):cmd_scan() Service found 0

Part of a tlogs.log possible log after successfully executed “scan” command

“info” command

Before the criminals can dispense cash, they first need to know the exact contents of the different cassettes. For this, they use the info command which provides exhaustive information on all cassettes and their contents. The list of used XFS API functions is the same as with the scan command, but this time WFSGetInfo is called with the WFS_INF_CDM_CASH_UNIT_INFO (303) constant passed as a param.

Below is an example of the data in log file returned by the info command.

(502):ExecuteCmd() Executing cmd
(506):ExecuteCmd() CMD = info
(402):cmd_info() ! hFoundGlobalService = 0
(213):GetDeviceInformation() ————————————————
(220):GetDeviceInformation() Device locked 0
(337):GetDeviceInformation() Module: C:\program files\dtatmw\bin\atmapp\atmapp.exe
Cash Unit # 1, name=SOMENAME
Type: 3
Status: HIGH
Currency ID: 0x52-0x55-0x42
Note Value: 5000
Notes Count: 3000
Notes Initial Count: 3000
Notes Minimum Count: 10
Notes Maximum Count: 0

Example5 Part of a tlogs.log possible log after successfully executed “info” command

“disp” command

The dispense command is followed by two additional params in the command file: currency and amount. Currency must contain one of the three-letter currency codes of notes kept in the CASH_UNIT_INFO structure (currency codes are described in ISO_4217 e.g. RUB, EUR). The amount code holds the amount of cash to dispense and this value must be a multiple of ten.

“die” command

Does nothing except deleting C:\ATM\c.ini command file.


ATMii is yet another example of how criminals can use legitimate proprietary libraries and a small piece of code to dispense money from an ATM. Some appropriate countermeasures against such attacks are default-deny policies and device control. The first measure prevents criminals from running their own code on the ATM’s internal PC, while the second measure will prevent them from connecting new devices, such as USB sticks.

The Festive Complexities of SIGINT-Capable Threat Actors

Wed, 10/04/2017 - 06:00

To read the full paper and learn more about this, refer to “Walking in Your Enemy’s Shadow: When Fourth-Party Collection Becomes Attribution Hell”

Attribution is complicated under the best of circumstances. Sparse attributory indicators and the possibility of overt manipulation have proven enough for many researchers to shy away from the attribution space. And yet, we haven’t even discussed the worst-case scenarios. What happens to our research methods when threat actors start hacking each other? What happens when threat actors leverage another’s seemingly closed-source toolkit? Or better yet, what if they open-source an entire suite to generate so much noise that they’ll never be heard?

Thankfully, the 2017 VirusBulletin conference is upon us and, as in previous years, we’re taking the opportunity to dive into an exciting subject, guided by our experience from doing hands-on APT research.

During the past years, we discussed the evolution of anti-malware research into intelligence brokerage, the inherent problems with doing attribution based solely on fifth-domain indicators, and an attempt to have a balanced discussion between defensive cats and the sly mice that elude them. Continuing in this direction, this year we decided to put our heads together to understand the implications that the esoteric SIGINT practice of fourth-party collection could have on threat intelligence research.

A few types of SIGINT Collection

The means by which information is generated and collected is the most important part of an analyst’s work. One must be well aware of the means and source of the information analyzed in order to either compensate or exploit its provenance. For that reason, collection can be categorized by its means of generation in relation to the position of the parties involved, as discussed below. These definitions will serve as functional categories for our understanding as outsiders looking into the more complex spheres of collection dynamics.

To showcase the types of data collection, let’s imagine a competent entity named ‘Agency-A’ as a stand-in for a ‘God on the wire‘-style SIGINT agency interested in fourth-party collection.

There are multiple types of collection categories available to this entity. The more obvious being information collected by Agency-A directly (first-party) or shared with Agency-A by partner services (second-party). Third-party collection, or information collected via access to strategic organizations, whether they realize it or not, has gotten a lot of attention over the past few years. This would include ISPs, ad networks, or social media platforms that aggregate great troves of valuable data.

Similarly, we will use further entities Agency-B as a second semi-competent SIGINT agency upon which Agency-A can be recurringly predatory for the sake of explanation. When necessary an even less competent Agency-C will serve as prey.

Yet, things get most interesting when we start talking about:

Fourth-party collection – …involves interception of a foreign intelligence service’s ‘computer network exploitation’ (CNE) activity in a variety of possible configurations. Given the nature of Agency-A as a cyber-capable SIGINT entity, two modes of fourth-party collection are available to it: passive and active. The former will take advantage of its existing visibility into data in transit either between hop points in the adversary’s infrastructure or perhaps in transit from the victim to the command-and-control servers themselves (whichever opportunity permits). On the other hand, active means involve the leveraging of diverse CNE capabilities to collect, replace, or disrupt the adversary’s campaign. Both present challenges we will explore in extensive detail further below.”

In less technical terms, fourth-party collection is the practice of spying on a spy spying on someone else. Or with age-old cryptographic interlocutors: Bob is obsessed with Alice. Alice is being spied on by her overzealous neighbour Eve. In order for Bob to be a creeper without arousing suspicion, he decides to spy on Eve with the purpose of getting to know Alice through Eve’s original privacy violation.

As you might expect there are different ways to do this and many of them enjoy the benefit of being near impossible to detect. Where possible, we have added examples of what to us looks like possible active attempts to collect on another’s collection. Otherwise, we have added thought experiments to help us wrap our heads around this shadowy practice. Two examples worth bringing to your attention (reproduced faithfully from our paper):

‘We heard you like popping boxes, so we popped your box so we can watch while you watch’

Attempting to highlight examples of fourth-party collection is a difficult exercise in the interpretation of shadows and vague remnants. While passive collection is beyond our ability to observe, active collection involves the risk of leaving a footprint in the form of artefacts. In the course of APT investigations, Kaspersky Lab’s Global Research and Analysis Team (GReAT) has encountered strange artefacts that defy immediate understanding in the context of the investigation itself. While we cannot be certain of the intent or provenance of these artefacts, they nonetheless fit a conceptual framework of active fourth-party collection. Here’s a few examples:

Crouching Yeti’s Pixelated Servers

In July 2014, we published our research on Crouching Yeti, also known as ‘Energetic Bear’, an APT actor active since at least 2010. Between 2010 and 2014, Crouching Yeti was involved in intrusions against a variety of sectors, including:

  • Industrial/machinery
  • Manufacturing
  • Pharmaceutical
  • Construction
  • Education
  • Information technology

Most of the victims we identified fell into the industrial and machine manufacturing sector, indicating vertical of special interest for this attacker.

To manage their victims, Crouching Yeti relied on a network of hacked websites which acted as command-and-control servers. For this, the attackers would install a PHP-based backend that could be used to collect data from or deliver commands to the victims. To manage the backend, the attackers used a control panel (also written in PHP) that, upon checking login credentials, would allow them to manage the information stolen from the victims.

In March 2014, while investigating one of the hacked sites used by Energetic Bear, we observed that for a brief period of time, the page for the control panel was modified to include an <img src> tag that pointed to a remote IP address in China. This remote 1×1 pixels wide image was likely intended to fingerprint the attackers as they logged into their control panel. The fingerprinting could have been used to collect attributory indicators. The usage of an IP address in China, which appeared to point to yet another hacked server, was most likely an attempt at a rudimentary false flag should this injection be discovered.

NetTraveler’s Most Leet Backdoor

While investigating the Nettraveler attacks, we obtained a disk image of a mothership server used by the threat actor. The mothership, a combination staging and relay server, contained a large number of scripts used by the attackers to interact with their malware, as well as VPN software and other IP masking solutions used to tunnel into their own hacking infrastructure.

Beyond the fortuitous boon of seizing such a content-rich server, GReAT researchers made a further unexpected discovery: the presence of a backdoor apparently placed by another entity.

We believe the backdoor was installed by an entity intent on maintaining prolonged access to the Nettraveler infrastructure or their stolen data. Considering that the NetTraveler operators had direct access to their mothership server and didn’t need a backdoor to operate it, we consider other possible interpretations less likely.

The artefact encountered is the following:

Name svchost.exe MD5 58a4d93d386736cb9843a267c7c3c10b Size 37,888

Interestingly, the backdoor is written in assembly and was injected into an empty Visual C executable that served as a template. This unusual implementation was likely chosen in order to confuse analysis or prevent detection by simple antivirus programs.

The backdoor is primitive and does nothing but listen to port 31337 (The most ‘LEET!’ port) and wait for a payload to be sent. The acceptable payload format is depicted here:

The assembly code is then executed and can perform any action chosen by the predatory attackers. The backdoor requires no authentication. Combining this sort of backdoor with Metasploit or other similar frameworks could have easily been used to control the system.

During the last years, we have seen a number of other peculiar incidents and cases which could constitute fourth party collection.”

To read the full paper and learn more about this, refer to “Walking in Your Enemy’s Shadow: When Fourth-Party Collection Becomes Attribution Hell”