From the Front Lines | Slam! Anatomy of a Publicly-Available Ransomware Builder

The barrier to entry for enterprising cybercriminals has been dropping considerably over recent years, in part due to the availability of RaaS (Ransomware as a Service) offerings on the darknet but also due to publicly-accessible code being shared for free. One such offering is the Slam Ransomware Builder, which had been hosted until recently on Github. In this post, we highlight how free ransomware builders like Slam offer an easy route into cybercrime and yet present a credible threat to organizations and enterprises. We provide a detailed list of indicators to help security teams detect and protect against Slam ransomware payloads.

Ransomware For “Educational Purposes Only”?

The Slam Ransomware Builder first appeared in late 2021, with Slam ransomware payloads appearing in the wild shortly after (e.g., ConsoleApp2.exe). During mid-2022, downloadable and executable versions of the Slam Ransomware Builder appeared on a publicly-visible repository on Github and were available for several months until Github admins removed the repository on September 1st, 2022.

The owner of the now-removed repository dubbed it “The Most Advanced Free Ransomware Builder” and has a history of providing “educational” videos on Vimeo, Youtube and KZHome, instructing viewers how to build ransomware and “virus payloads”.

Slam ransomware builder video hosted on Vimeo
Source: Slam ransomware builder video hosted on Vimeo

While the author’s public postings contain the usual “for educational purposes only” and “don’t try this” disclaimers to avoid responsibility, they also contain language such as “most advanced ransomware” and “damage rate: destructive”.

Slam ransomware builder video hosted on Youtube
Source: Slam ransomware builder video hosted on Youtube

The author had described the ransomware’s behavior in detail in earlier publicly-posted videos, describing how victim data could be exfiltrated to an attacker-controlled site.

The author’s reasons for distributing free ransomware builders can only be guessed at, but despite being free, the builder and payloads are genuine threats that can cause real damage. As our analysis below shows, Slam is a full-featured ransomware with AES256 encryption, UAC bypass, shadow backup copy deletion and data exfiltration capabilities. In other words, everything needed to lock and steal enterprise data.

Slam Ransomware Builder Features

The most recent release of the Slam ransomware builder prior to being removed from Github was version 1.7. Earlier versions of the tool supported either English or Spanish locales, while later versions including 1.7 allow toggling between the two.

The existing feature set includes the following:

  • Fully customizable ransom notes
  • Custom encryption passphrases
  • All ransomware to lay dormant until a network is available
  • UAC Bypass (1)
  • Run external commands with the ransomware launch
  • VSS/ backup deletion
  • Basic file transfers (HTTP) for exfiltration

Despite the code being removed from Github, it is possible the author intends to find or already has other distribution outlets. A list of features promised for the future include screen locking, MBR overwrites, and “LogonUI overwriting”.

Upon running the code provided on Github, users of the builder are presented with a menu leading to different builder components or indications of their upcoming release.

Version 1.6 of the Slam Ransomware Builder
Version 1.6 of the Slam Ransomware Builder

When choosing the “slam ransomware builder” option, users must first “Install”, then “Start” to launch the builder interface. This installation essentially consists of writing the builder EXE to c:slam_ransomware_builder. Any other component requiring an “Install” step will also go to the root of the C drive (e.g., c:slam_mbr_builder)

Once the main interface is launched, the user is presented with a standard set of options for building their ransomware payloads.

Options including the following are present in this interface:

  • Ransom note name and text
  • Wallpaper modification options and images
  • Affected file extensions
  • File encryption (types / extensions to encrypt)
  • Remote folder options (OneDrive)

The tool provides more ‘Advanced’ configuration options as well. These options are accessible via the “advanced” button.

Options in this section include:

  • Network awareness (remain idle until Wi-Fi is available)
  • Verbose output options (decrypter)
  • Persistence (add to startup)
  • Inhibit recovery (website blocking, self-destruction, backup destruction).

The “block antivirus websites” option is meant to inhibit the victims from being able to download security software or check suspicious files on public malware repository sites such as VirusTotal.

The ransomware achieves this by modifying the device’s Hosts file, adding a long list of sites belonging to the likes of Avast, Avira, Bitdefender, CCleaner, Google, Kaspersky, McAfee, Microsoft, Panda Security, Trend Micro, VirusTotal, YouTube, and others. Each site is simply bound to the machine’s loopback address (typically, 127.0.0.1), preventing the domain name from being resolved to an external IP address.

Some of the almost 100 domain names added to the Hosts file
Some of the almost 100 domain names added to the Hosts file

With regard to bypasses, the version of Slam we analyzed includes a single UAC bypass, based on UACMe, which attempts to defeat Windows User Account Control by abusing the built-in Windows AutoElevate backdoor. UACMe is a bypass technique that has been known for some years and widely abused by a number of other malware families including Multiplug adware, Dyre, Empercrypt and IcedID.

To exfiltrate victim data, the user can specify an HTTP server in the configuration interface, where a connection test can also be performed. If the connection test fails, an error is displayed. Other options available to the user include USB infection and execution of custom commands when the payload is detonated on the victim machine.

Slam Ransomware Payloads

With all options configured, the executable payloads generated are standard EXE files. The builder outputs both the encryptor and decryptor tools.

When executed with non-Administrator privileges, the UAC prompts and/or configured bypasses will come into play.

Slam payload UAC prompt
Slam payload UAC prompt

Post-execution, the victim device is encrypted according to the options configured in the builder.

The payload is written to %AppData%Localdiscord.exe, which is called in the registry (Run key), ensuring the ransomware payload is persistent.

As advertised, the Slam payload successfully inhibits recovery via removal of VSS backups on an unprotected machine. Both wmic and vssadmin methods are utilized for VSS deletion.

/c vssadmin delete shadows /all /quiet & wmic shadowcopy delete & bcdedit /set {default} bootstatuspolicy ignoreallfailures & bcdedit /set {default} recoveryenabled no & wbadmin delete catalog -quiet

wmic shadowcopy delete

The ransomware also deletes various logs, Windows installation and recovery-related files via cleanmgr.exe. In the payload we analyzed, for example, a process named wgMHhFHnkiczPUNfqaA8Cx4kqwVcRG.exe issues the cleanmgr.exe command with the /AUTOCLEAN parameter, which executes Windows disk cleanup and removes Windows installation files on unprotected devices.

system32cleanmgr.exe /autoclean /d C:

Slam MBR Builder

The Slam builder also contains a very early stage “Alpha” MBR builder tool. Choosing to “Install” should write start.exe to c:slam_mbr_builderstart.exe. This does not appear to occur in our testing and analysis, and the feature appears to be non-functional in the version of the Slam Builder we analyzed from Github.

However, we were able to obtain a copy of the builder from another source that allowed us to launch the builder and observe the output.

Slam “Alpha” MBR builder

Within the MBR Builder interface, users are able to configure the message displayed to the victim.

Slam MBR Builder Ransom Note Configuration
Slam MBR Builder Ransom Note Configuration

Prior to executing the build, a final screen allows the attacker to choose the “reboot mode” with the options being

  • Do Nothing
  • BSOD
  • Reboot
  • Shutdown
  • Nothing

Payloads from the MBR builder have been observed in the wild with the following PDB string.

C:slam_mbr_builderMbrOverwritermbrcsobjDebugmbrcs.pdb

Conclusion

In this area and many others of infosec, there is a fine line between “education” and researcher-led offensive security that seeks to explore and improve weaknesses in enterprise defenses on the one hand, and simple, out-and-out malicious code designed to aid and abet criminal offenses on the other. We see no indications in the various public artifacts around the Slam ransomware builder (code, videos, Github repository) that suggest it could reasonably be interpreted as in the service of the former.

However that may be, once in the hands of unscrupulous actors, full-featured projects such as these represent a real risk to enterprises and organizations.

We applaud Github for removing this code and hope this post serves as a reminder to defenders to be vigilant as threat actors continue to simplify the ransomware-centric extortion process. The barrier to entry into the world of cybercrime has never been lower.

SentinelOne Singularity™ detects and prevents malicious behavior associated with Slam Ransomware and its associated artifacts.

Indicators of Compromise

Observed File Names
ConsoleApp2.exe
mbrcs.exe
JpegMedic ARWE
slam ransomware builder.exe

Observed PDB Strings

C:slam_mbr_builderMbrOverwritermbrcsobjDebugmbrcs.pdb
c:slam_ransomware_builderConsoleApp2ConsoleApp2objDebugConsoleApp2.pdb
C:UsersamdgaDesktopUACME-masterSourceAkagioutputWin32DebugAkagi.pdb
D:agent_work20sbinariesx86retbini386vcruntime140d.i386.pdb
c:slam_ransomware_builderuacConsoleApp2objDebugConsoleApp2.pdb
c:slam_ransomware_builderConsoleApp2ConsoleApp2objDebugConsoleApp2.pdb
C:slam_mbr_builderMbrOverwritermbrcsobjDebugmbrcs.pdb
C:UsersamdgasourcereposconectconectobjDebugconect.pdb
C:Usersandersourcereposslam ransomware builderslam ransomware builderobjDebugslam ransomware builder.pdb

SHA1 Hashes
1ba9043ac164c6c60de4a1ee2ca50b2e7f4ebaf5
2037d9f2e7cd15930e83f5142c5a48adecd3b617
272566e8b5880e32cefb7a165a833652815a003f
27b1ca0793caa19edabfbc49e6cffc05b73093da
2c41f64557056e69541acf5ba52313869122f625
336371f4200af680f73c0b9c51fca5a25dd5754a
35ab1d4924990bf98a8e2e1026f91b5c9052de8e
3fa6705ca1b056a66f25a689dff72af0893f5b86
40bfa92e86484c09f2f7668121a1c4047c17ae72
44aaef83b79f4e963c4fee56250bc053eae5ec64
4879bd193dd73681c977371c857217257f141c92
4cff2b02cb6c1f866499125c003af1032a81b480
5a28f787cc73cffa7b5786faf3298d43e00d12aa
61e8ba86725ec3f4e034c51950cabc6254c5cca5
6325c42719b1aa3a48dd39b8add200054d3e0118
669ce00937bde782a88526205f083861e6d71be1
6e420a6c7b8e2d144df66dcbbae1afba62c82f4b
7429fdf9151dfa9e4d4dc8ef86528313d13dc73f
7690c273c8164a65602ed8f4284f0d50966d27c6
863edff3c71e89349674df35ab07f27ecb6702ef
880c343e75e7e8731f185ce756357599c37be065
8b46ce2ffa24a377ff30ea094e02bc3ba3e808da
8f3dc8437563182e06699763581fd6f7923b7582
9edd3d920fbe89240d52cc8b300a90e5bf576f73
b031d4c3747b58d930f33fe73abbf518dac63a31
be82474f54f49249c43c701c12907ec730e2a723
c5351846988ef5d6e7b95f564416138f59e2092a
c84aeb8c0b3939fd7f6beb9d73e72cc5ed8745db
c998384c7b8cfd2ca881f282dfdbc104d8402bac
ca2999c9c5a17b0253579194f651b4aafdce16f1
cb243b61a8d43816e1de7f0767b1377d0276dd71
cf30cc1e653043df81aa9d8974f2f927ceadc826
d187d81f4d021839e8f6e925dc192e231eb4679c
d635103117daaf2a2b93d465e32e7b722dd4d367
d6c9a556f5770f0a8f8ad05c5d46becd0cd021d3
d94eb94bb3c2c6c0c70916f8be2417ac616e8b43
dc327f3afbb6c770656be16fc885e1090f8395a3
ddba71aae3b8139210f71e835e1b89e90b0bd1dc
e0868fdb2f09d3a4aefe4c79d6af88c2f9b55ce2
e2052995d368355e899a518dbbbab716045abbd1
e9a5b40d0ba5a8bb5c4a1c5471616c93e0851558
ea4f7dda5a64a740a9c5570870ccba2788c69ea6
ee144154139619b8c1d890e5b6f9bf130d929e6f
eeafbbfaaf05d8b7a8a1dc3f7858a21e7fdb0531
f31855a1d5509b1e906caee75db3326515488cbc
fcd90af249796fc3c40e1e94d558b6f2d61304b5

MITRE ATT&CK
T1542.003 – Pre-OS Boot: Bootkit
T1047 – Windows Management Instrumentation
T1547.001 – Boot or Logon Autostart Execution: Registry Run Keys / Startup Folder
T1564.003 – Hide Artifacts: Hidden Window
T1112 – Modify Registry
T1490 – Inhibit System Recovery
T1486 – Data Encrypted for Impact
T1491.001 – Defacement: Internal Defacement
T1083 – File and Directory Discovery
T1005 – Data from Local System
T0809 – Data Destruction

Say Hello to Crazy Thin ‘Deep Insert’ ATM Skimmers

A number of financial institutions in and around New York City are dealing with a rash of super-thin “deep insert” skimming devices designed to fit inside the mouth of an ATM’s card acceptance slot. The card skimmers are paired with tiny pinhole cameras that are cleverly disguised as part of the cash machine. Here’s a look at some of the more sophisticated deep insert skimmer technology that fraud investigators have recently found in the wild.

This ultra thin and flexible “deep insert” skimmer recently recovered from an NCR cash machine in New York is about half the height of a U.S. dime. The large yellow rectangle is a battery. Image: KrebsOnSecurity.com.

The insert skimmer pictured above is approximately .68 millimeters tall. This leaves more than enough space to accommodate most payment cards (~.54 mm) without interrupting the machine’s ability to grab and return the customer’s card. For comparison, this flexible skimmer is about half the height of a U.S. dime (1.35 mm).

These skimmers do not attempt to siphon chip-card data or transactions, but rather are after the cardholder data still stored in plain text on the magnetic stripe on the back of most payment cards issued to Americans.

Here’s what the other side of that insert skimmer looks like:

The other side of the deep insert skimmer. Image: KrebsOnSecurity.com.

The thieves who designed this skimmer were after the magnetic stripe data and the customer’s 4-digit personal identification number (PIN). With those two pieces of data, the crooks can then clone payment cards and use them to siphon money from victim accounts at other ATMs.

To steal PINs, the fraudsters in this case embedded pinhole cameras in a false panel made to fit snugly over the cash machine enclosure on one side of the PIN pad.

Pinhole cameras were hidden in these false side panels glued to one side of the ATM, and angled toward the PIN pad. Image: KrebsOnSecurity.com.

The skimming devices pictured above were pulled from a brand of ATMs made by NCR called the NCR SelfServ 84 Walk-Up. In January 2022, NCR produced a report on motorized deep insert skimmers, which offers a closer look at other insert skimmers found targeting this same line of ATMs.

Image: NCR

Here are some variations on deep insert skimmers NCR found in recent investigations:

Variations on deep insert skimmers recently found inside compromised ATMs.

The image on the left below shows another deep insert skimmer and its constituent components. The picture on the right shows a battery-operated pinhole camera hidden in a false fascia directly to the right of the ATM’s PIN pad.

Images: NCR.

The NCR report included additional photos that show how fake ATM side panels with the hidden cameras are carefully crafted to slip over top of the real ATM side panels.

Image: NCR.

Sometimes the skimmer thieves embed their pinhole spy cameras in fake panels directly above the PIN pad, as in these recent attacks targeting a similar NCR model:

Image: NCR

In the image below, the thieves hid their pinhole camera in a “consumer awareness mirror” placed directly above an ATM retrofitted with an insert skimmer:

Image: NCR

The financial institution that shared the images above said it has seen success in stopping most of these insert skimmer attacks by incorporating a solution that NCR sells called an “insert kit,” which stops current skimmer designs from locating and locking into the card reader. NCR also is conducting field trials on a “smart detect kit” that adds a standard USB camera to view the internal card reader area, and uses image recognition software to identify any fraudulent device inside the reader.

Skimming devices will continue to mature in miniaturization and stealth as long as payment cards continue to hold cardholder data in plain text on a magnetic stripe. It may seem silly that we’ve spent years rolling out more tamper- and clone-proof chip-based payment cards, only to undermine this advance in the name of backwards compatibility. However, there are a great many smaller businesses in the United States that still rely on being able to swipe the customer’s card.

Many newer ATM models, including the NCR SelfServ referenced throughout this post, now include contactless capability, meaning customers no longer need to insert their ATM card anywhere: They can instead just tap their smart card against the wireless indicator to the left of the card acceptance slot (and right below the “Use Mobile Device Here” sign on the ATM).

For simple ease-of-use reasons, this contactless feature is now increasingly prevalent at drive-thru ATMs. If your payment card supports contactless technology, you will notice a wireless signal icon printed somewhere on the card — most likely on the back. ATMs with contactless capabilities also feature this same wireless icon.

Once you become aware of ATM skimmers, it’s difficult to use a cash machine without also tugging on parts of it to make sure nothing comes off. But the truth is you probably have a better chance of getting physically mugged after withdrawing cash than you do encountering a skimmer in real life.

So keep your wits about you when you’re at the ATM, and avoid dodgy-looking and standalone cash machines in low-lit areas, if possible. When possible, stick to ATMs that are physically installed at a bank. And be especially vigilant when withdrawing cash on the weekends; thieves tend to install skimming devices on Saturdays after business hours — when they know the bank won’t be open again for more than 24 hours.

Lastly but most importantly, covering the PIN pad with your hand defeats one key component of most skimmer scams: The spy camera that thieves typically hide somewhere on or near the compromised ATM to capture customers entering their PINs.

Shockingly, few people bother to take this simple, effective step. Or at least, that’s what KrebsOnSecurity found in this skimmer tale from 2012, wherein we obtained hours worth of video seized from two ATM skimming operations and saw customer after customer walk up, insert their cards and punch in their digits — all in the clear.

If you enjoyed this story, check out these related posts:

Crooks Go Deep With Deep Insert Skimmers

Dumping Data from Deep Insert Skimmers

How Cyber Sleuths Cracked an ATM Shimmer Gang

Endpoint, Identity and Cloud | Top Cyber Attacks of 2022 (So Far)

Businesses have thrived in the era of more – more tools, more access, and more connections. When it comes to the digital landscape though, the notion of having more doesn’t come without risk. While businesses have continued to grow and scale, cyber attacks have done the same and quickly so.

This post discusses some of the most dangerous endpoint, identity, and cloud-based cyber attacks from the first three quarters of this year. Understanding their causes and impacts is the first step businesses can take to strengthen their defenses against similar attacks in the future.

Endpoint-Based Attacks

Endpoint attacks have evolved in the last two decades from computer viruses to sophisticated ransomware campaigns targeting high-profile organizations. The challenge today is that protecting endpoints isn’t what it used to be. Threat actors are professionalizing, turning ransomware activities into full-scale service models. The rise of Ransomware-as-a-Service (RaaS) means that even low-level cybercriminals can now access and become profitable off of complex malware tools.

In the first three quarters of this year alone, ransomware has targeted multiple critical infrastructure organizations, including those listed below:

  • Bernalillo County (Ransomware attack) – In January 2022, New Mexico’s most heavily populated county, experienced a ransomware attack which took out many of its government systems and services. The attack forced officials to keep most county buildings closed to the public.
  • Denso (Pandora Ransomware) – In February, Denso, a Fortune 500 company supplying automotive parts for Toyota, Honda, Mercedes-Benz, Volvo, Fiat, General Motors, and Ford, detected unauthorized, third-party access to its network. The attack was later claimed by ransomware gang Pandora, who then threatened to leak 1.4 terabytes of the company’s trade secrets and transactional records.
  • Bridgestone (LockBit Ransomware) – In March, Bridgestone was hit with a cyber attack that forced operators to shut down affected computer networks and production across its North American factories. The LockBit ransomware group later took responsibility for this attack.
  • Costa Rican Government (Conti) – In April, a ransomware attack on the Costa Rican government led to the first national emergency declared in response to a cyber attack. The impact of the ransomware attack affected government services, the country’s ministry of finance, as well as the import and export sectors. Later claimed by Conti ransomware group, the government was asked to pay $20 million dollars. Shortly after, Conti group hit the Costa Rican government a second time, this time using HIVE ransomware to cause widespread disruption of the country’s public health services systems.
  • Spice Jet (Ransomware attack) – India’s second largest airline, SpiceJet, faced a ransomware attack in May, leading to a cascading delay of flights which stranded many passengers at both airports and within aircrafts. Many passengers aired their frustrations regarding the delay and lack of communications over social media.
  • Entrust (LockBit 3.0) – In June, Digital security firm, Entrust, confirmed that its networks were breached by a ransomware gang who successfully stole data from their internal systems. Entrust’s services include identity management, comms encryption, and secure digital payments making news of their ransomware attack an immediate concern for organizations using their software for authentication. Subsequently, Entrust was found to be added to the LockBit 3.0 Tor-based website.
  • Knauf (Black Basta) – In July, Black Basta ransomware gang claimed responsibility for their cyber attack on Knauf, the multinational building and construction materials giant. Knauf’s global team was forced to shut down all of their IT systems to isolate the attack, which disrupted business operations and delivery processes. Post-attack, Black Basta published 20% of the exfiltrated files, congruent to their notoriety for double-extorting their high-profile victims.

In addition, in the 3rd quarter of 2022, CISA and the FBI warned of a number of ongoing, widespread ransomware campaigns currently attacking unnamed businesses and organizations.

  • Zeppelin Ransomware Campaign – In August, the FBI and CIS released a joint cybersecurity advisory to share known indicators of compromise, as well as tactics, techniques, and procedures of Zeppelin malware, functioning as a RaaS. This malware has been used against defense contractors, educational institutions, manufacturers, technology companies, and especially organizations in the healthcare and medical industries.
  • Vice Society Campaign – In September, the FBI, CISA, and Multi-State Information Sharing and Analysis Center (MS-ISAC) released a joint cybersecurity advisory warning against a disproportionate increase of ransomware attacks on the education sector by Vice Society in tandem with the start of the 2022-2023 academic year for American schools.

Identity-Based Attacks

Ransomware, data breaches, and supply chain attacks saturate global news headlines, but another rising threat has gained traction in 2022. Identity-based attacks are now a threat businesses keep at the forefront of their threat awareness efforts. With remote workforces, widespread adoption of IoT, and the huge numbers of digital identities being created even for a single organization, the attack surface continues to widen, leaving businesses vulnerable to identity-based exploitation by opportunistic threat actors.

Attacks on Active Directory – Cisco

Too often, threat actors weaponize legitimate tools and solutions that their targets use. Active Directory (AD) works by storing information about objects on a network in a logic hierarchy to make information easy to find for administrators and users. As seen in several identity-based attacks over the last few quarters, threat actors leverage Active Directory (AD) infrastructure in their ransomware campaigns and extortion efforts especially when there is a lack of identity protection. Consider the following examples where ransomware gangs targeted AD as part of their tactics.

In late 2021, researchers reported on a recent BazarLoader infection and how it led to the use of Cobalt Strike, and finally Conti ransomware to perform network reconnaissance. Just three minutes after the initial compromise, the threat actor used ADFind, a command line tool, to enumerate an AD environment on the infected host. By compromising AD, the actors were able to discover users, computers, file shares, and more from the environment. Typically, a threat actor’s next step is to gain access to the domain controller and other network servers, moving laterally into the system.

The Cisco breach that occurred in May 2022 leveraged legitimate employee credentials synced in an employee’s browser and a combination of vishing (voice phishing) attacks and MFA fatigue techniques to achieve VPN access to the targeted network. Once in, the threat actor exfiltrated the contents of a Box folder and the employee’s authentication data from Active Directory.

Abuse of AD serves threat actors well as it is designed to provide convenient access into a network. Compromising AD means threat actors can move deep into the network, escalating their access rights and encrypting andexfiltrating data on the way. With AD being the crown jewels of a business, attackers have zeroed in on targeting identity and access management gaps to reach what it is they want.

Attacks on Identity Management Platforms – Okta and Lapsus$

In March of this year, Lapsus$ digital extortion gang published what looked like substantial amounts of source code from Microsoft’s Bing and Cortana products. Though a potential Microsoft breach was serious enough, Lapsus$ also posted screenshots of their control over an Okta super admin account. Okta is a popular identity management platform used by thousands of large-scale organizations allowing users to access multiple services and apps through a single login interface.

Lapsus$’s control of an Okta super admin account is dire indeed as businesses increasingly rely on identity management software to streamline login experiences for their employees, partners, and customers. Businesses are falling victim to more account takeovers that directly stem from compromised identity management vendors, giving threat actors system privileges such as resetting account passwords, changing account email addresses, and access to sensitive data.

As ransomware and other malicious actors target on-premises Active Directory and cloud-hosted Azure AD for initial access and lateral movement, Identity protection has become a must for organizations.

Cloud-Based Attacks

The accelerated move from on-prem to hybrid and cloud environments has introduced a pressing need for businesses to keep their cloud workloads safe from threat actors. Cloud servers allow businesses to scale with ease, boosting efficiency, but also requires unique considerations such as securing serverless workloads and Kubernetes, virtual machines, and containers.

Amazon Web Services (AWS)

A subsidiary of Amazon, AWS is a comprehensive cloud computing platform providing a variety of on-demand services such as data storage, content delivery, networking, and more. One of its main services is Amazon Simple Storage Service (S3) – an object storage service built to house and retrieve any amount of data for its users. Objects (files) are then stored in S3 buckets which serve as containers for any amount of data belonging to an account.

While AWS S3 buckets are highly popular, they have become a prime target for threat actors as they are accessible to the public and are often misconfigured. Once an S3 bucket is compromised, it provides the threat actor with access to incredible amounts of data of which they could exfiltrate, use for ransom, sell on darknet marketplaces, or all of the above.

In the recent Civicom data leak, the misconfiguration of an S3 bucket resulted in a massive data leak, compromising over 100,000 files. In this case, the bucket was left open without password or security verification. The online video conferencing firm reported that 8 terabytes of stolen data included the video and audio files of customers’ meetings, recordings, and transcripts. As the firm’s main customer base included B2B companies, much of the data may have contained private company secrets or intellectual property. Further, the leak also revealed personally identifiable information (PII) of many of Civicom’s own employees.

The July breach of Pegasus Airlines showcases yet another example of unprotected S3 buckets leading to data loss. In this attack, the airline reported 6.5 terabytes of data was compromised with over 23 million files publicly exposed. Files in the unprotected bucket were linked to proprietary software developed by the company for use in aircraft navigation and in-flight processes such as take off and landing, refueling, and safety procedures. Pegasus Airlines also confirmed that sensitive information such as the PII of flight crews, source code, secret keys, and even plain-text passwords were also exposed. At least two other affiliated airlines using the same proprietary software may also be compromised in relation to this breach, exponentially increasing the number of total persons affected.

Kubernetes Vulnerabilities

Kubernetes is an open-source system that automates the deployment, scaling, and management of applications running in containers. It uses a cluster architecture composed of many control plans and one of more virtual or physical machines called worker nodes. The worker nodes are what host “Pods” – components of the application workload. The control plane exists to establish policy which manages the worker nodes and Pods in the cluster. Since the control plan is responsible for running across multiple endpoints to provide fault-tolerance and high availability, it is a valuable target for threat actors seeking to leverage its infrastructure for malicious purposes or to cause a denial of service attack.

As it is hosted in a cloud environment, Kubernetes is afflicted with the same main threat vectors that clouds are susceptible to:

  • Supply Chain Risks – These kinds of risks can occur at the container level if a malicious container or third-party application provides threat actors with a foothold in the cluster. Actors could also gain a foothold into any of the worker nodes or part of the control plane was compromised.
  • Malicious Threat Actors – Threat actors can exploit vulnerabilities and misconfigurations in components of the Kubernetes infrastructure allowing them to gain access from a remote location.
  • Insider Threats – Administrators, users, or cloud service providers (CSPs) would all have access to physical systems or hypervisors managing Kubernetes nodes. This level of access could be used to compromise a Kubernetes environment.

How SentinelOne Measures Up to 2022 Cyber Attacks

2022 has, so far, been a complex year as businesses settle back into offices and hybrid workspaces but face the ramifications of geopolitical uncertainty, economic downturn, and cyber attacks that are climbing to new heights. Having more tools, access, and connections has no doubt benefited businesses, but it has also opened up a larger attack surface in which threat actors can operate.

While no business is immune from cyber attacks, examining the most dangerous attacks of the first three quarters of 2022 allows for better preparation for the following quarter and beyond. SentinelOne’s autonomous, AI-driven solutions can help deliver comprehensive security for those in search of endpoint, identity, and cloud protection.

In a single cybersecurity platform, Singularity XDR, fuses together the data, access, control, and integration planes of its endpoint protection (EPP), endpoint detection and response (EDR), IoT security, cloud workload protection (CWPP), and identity threat detection and response (ITDR). With Singularity, organizations gain access to back-end data across the organization through a single solution, providing a cohesive view of their network and assets by adding a real time autonomous security layer across all enterprise assets.

Request a demo of Singularity XDR to start leveraging AI-powered prevention, detection, response, and threat hunting across user endpoints, containers, cloud workloads, and IoT devices. Need expert advice? Contact us here.

SentinelOne Singularity XDR
Supercharge. Fortify. Automate. Extend protection with unfettered visibility, proven protection, and unparalleled response. Discover the power of autonomous with Singularity XDR.

Wormable Flaw, 0days Lead Sept. 2022 Patch Tuesday

This month’s Patch Tuesday offers a little something for everyone, including security updates for a zero-day flaw in Microsoft Windows that is under active attack, and another Windows weakness experts say could be used to power a fast-spreading computer worm. Also, Apple has also quashed a pair of zero-day bugs affecting certain macOS and iOS users, and released iOS 16, which offers a new privacy and security feature called “Lockdown Mode.” And Adobe axed 63 vulnerabilities in a range of products.

Microsoft today released software patches to plug at least 64 security holes in Windows and related products. Worst in terms of outright scariness is CVE-2022-37969, which is a “privilege escalation” weakness in the Windows Common Log File System Driver that allows attackers to gain SYSTEM-level privileges on a vulnerable host. Microsoft says this flaw is already being exploited in the wild.

Kevin Breen, director of cyber threat research at Immersive Labs, said any vulnerability that is actively targeted by attackers in the wild must be put to the top of any patching list.

“Not to be fooled by its relatively low CVSS score of 7.8, privilege escalation vulnerabilities are often highly sought after by cyber attackers,” Breen said. “Once an attacker has managed to gain a foothold on a victim’s system, one of their first actions will be to gain a higher level of permissions, allowing the attacker to disable security applications and any device monitoring. There is no known workaround to date, so patching is the only effective mitigation.”

Satnam Narang at Tenable said CVE-2022-24521 — a similar vulnerability in the same Windows log file component — was patched earlier this year as part of Microsoft’s April Patch Tuesday release and was also exploited in the wild.

“CVE-2022-37969 was disclosed by several groups, though it’s unclear if CVE-2022-37969 is a patch-bypass for CVE-2022-24521 at this point,” Narang said.

Another vulnerability Microsoft patched this month — CVE-2022-35803 — also seems to be related to the same Windows log file component. While there are no indications CVE-2022-35803 is being actively exploited, Microsoft suggests that exploitation of this flaw is more likely than not.

Trend Micro’s Dustin Childs called attention to CVE-2022-34718, a remote code execution flaw in the Windows TCP/IP service that could allow an unauthenticated attacker to execute code with elevated privileges on affected systems without user interaction.

“That officially puts it into the ‘wormable’ category and earns it a CVSS rating of 9.8,” Childs said. “However, only systems with IPv6 enabled and IPSec configured are vulnerable. While good news for some, if you’re using IPv6 (as many are), you’re probably running IPSec as well. Definitely test and deploy this update quickly.”

Cisco Talos warns about four critical vulnerabilities fixed this month — CVE-2022-34721 and CVE-2022-34722 — which have severity scores of 9.8, though they are “less likely” to be exploited, according to Microsoft.

“These are remote code execution vulnerabilities in the Windows Internet Key Exchange protocol that could be triggered if an attacker sends a specially crafted IP packet,” wrote Jon Munshaw and Asheer Malhotra. “Two other critical vulnerabilities, CVE-2022-35805 and CVE-2022-34700 exist in on-premises instances of Microsoft Dynamics 365. An authenticated attacker could exploit these vulnerabilities to run a specially crafted trusted solution package and execute arbitrary SQL commands. The attacker could escalate their privileges further and execute commands as the database owner.”

Not to be outdone, Apple fixed at least two zero-day vulnerabilities when it released updates for iOS, iPadOS, macOS and Safari. CVE-2022-32984 is a problem in the deepest recesses of the operating system (the kernel). Apple pushed an emergency update for a related zero-day last month in CVE-2022-32983, which could be used to foist malware on iPhones, iPads and Macs that visited a booby-trapped website.

Also listed under active attack is CVE-2022-32817, which has been fixed on macOS 12.6 (Monterey), macOS 11.7 (Big Sur), iOS 15.7 and iPadOS 15.7, and iOS 16. The same vulnerability was fixed in Apple Watch in July 2022, and credits Xinru Chi of Japanese cybersecurity firm Pangu Lab.

“Interestingly, this CVE is also listed in the advisory for iOS 16, but it is not called out as being under active exploit for that flavor of the OS,” Trend Micro’s Childs noted. “Apple does state in its iOS 16 advisory that ‘Additional CVE entries to be added soon.’ It’s possible other bugs could also impact this version of the OS. Either way, it’s time to update your Apple devices.”

Apple’s iOS 16 includes two new security and privacy features — Lockdown Mode and Safety Check. Wired.com describes Safety Check as a feature for users who are at risk for, or currently experiencing, domestic abuse.

“The tool centralizes a number of controls in one place to make it easier for users to manage and revoke access to their location data and reset privacy-related permissions,” wrote Lily Hay Newman.

“Lockdown Mode, on the other hand, is meant for users who potentially face targeted spyware attacks and aggressive state-backed hacking. The feature comprehensively restricts any nonessential iOS features so there are as few potential points of entry to a device as possible. As more governments and repressive entities around the world have begun purchasing powerful commodity spyware to target individuals of particular importance or interest, iOS’s general security defenses haven’t been able to keep pace with these specialized threats.”

To turn on Lockdown Mode in iOS 16, go to Settings, then Privacy and Security, then Lockdown Mode. Safety Check is located in the same area.

Finally, Adobe released seven patches addressing 63 security holes in Adobe Experience Manager, Bridge, InDesign, Photoshop, InCopy, Animate, and Illustrator. More on those updates is here.

Don’t forget to back up your data and/or system before applying any security updates. If you experience glitches or problems installing any of these patches this month, please consider leaving a comment about it below; there’s a decent chance other readers have experienced the same and may chime in here with useful tips.

Fortune Names SentinelOne a Top Workplace in Tech 

SentinelOne has been named a Best Workplace in Technology on Fortune’s US Ranking for 2022. Over 150K employees working in the tech industry were surveyed on the quality and consistency of the employee experience on attributes including trust, respect, credibility, fairness, pride, and camaraderie.

We spoke to a few teammates to learn more about what makes our organization a top workplace. Tremendous pride in our leading-edge deep tech, transformative professional and personal opportunities and the equitable culture at SentinelOne were at the top of the list.

David Jung, Staff Software Engineer

David is building his career at SentinelOne and loving his fully-remote New York City lifestyle. He started at SentinelOne a little over three years ago as an intern while studying at the University of British Columbia.

“Working at SentinelOne while I was studying Computer Engineering was life-changing,” said David, who was also a Sentinel during the record-breaking IPO. “There was so much energy in the room the day we went public. There was an intense air of celebration, and then equally as intense – we were ready to get back to work!”

David has enjoyed significant career growth since starting at SentinelOne as an intern. He recently received his third promotion, going from Engineer, to Senior Engineer to now Staff Engineer.


“I have progressed very quickly by working on amazing tech with amazing people. They are all extremely smart, and at the same time, very humble. That’s a great combination.”

David works on the event database team, migrating customers into our systems and querying endpoint detection and response events. This includes developing new features, as well as making our system more scalable and efficient.

Our goal is to ingest and query as much data as fast as we possibly can,” said David. “The amount that we are dealing with solves a lot of interesting engineering problems and that’s what I like about my job.

When David is not working, he enjoys biking, running and spending time with his dog Spitzer, a Beagle Terrier mix. He recently picked up a new hobby that allows him to express his creativity.

“I love writing code, but I can’t physically touch it,” said David. “I recently took a 12-week pottery class where I learned how to make bowls, vases and pots. It’s great to make something I can hold onto.”

David’s remote team stays connected on Zoom and Slack. His team recently had an offsite, where they were grateful to connect in person. He also enjoys spending time at the local WeWork.

“My place isn’t that big living in New York City,” said David. “It’s great having the perk to be able to enjoy really cool office space.”

Elena Militsyna, Senior Software Engineer

Writing code is Elena’s favorite part of her job. She’s been working at SentinelOne on the Cloud Security Team for just four months and is proud to be part of such a talented team.


“It’s so cool building something from scratch to help so many customers,” said Elena. “The tech itself is an amazing  instrument, but it’s our people who make it all work so well.”

Elena describes our deep technology as the best on the market.

“There’s not many companies that can offer the opportunity to work on state-of-the-art technology,” said Elena. “We are offering solutions that provide direct value to customers by keeping all of their data safe in one autonomous platform. And we just keep getting better!

Elena said she is also grateful for the focus on wellness and career growth in our workplace culture. Prior to working here, she worked as Java developer for several companies, including a large corporation and a smaller machine learning startup, and had a chance to compare working environments of a different style. She believes SentinelOne is combining the advantages of a corporation with established processes and the innovative vibe and ability to change quickly of a startup.

Here we have great work-life balance,” said Elena. “We have a very fast pace when we innovate. There can be stress at times, but it’s always balanced. Our team is growing, and the company is giving us the resources we need to grow ourselves and do our best work.”

Elena works in the newly-opened Prague Office in the Czech Republic in Missouri Park with a view of Prague Castle. Built with sustainability in mind, the location has Leed Platinum Status, the highest level of certification for green buildings.

“We love going into the office, it’s so fun to work there,” said Elena. “People go in when they can, they want to be together.”

Elena immigrated to the Czech Republic four years ago, from a small town in Russia. She studied complex protection of information systems at Orenburg State University.

“My father encouraged me to study technical things,” Elena said. “Even before I started working in cybersecurity, I studied it for five years. It’s fascinating to me.”

Noa Frankel, Senior Product Manager

Noa is part of a cross-functional global team working on the Singularity XDR Platform, providing customers with a limitless data platform to ingest, retain, correlate, search and action all enterprise security data – real time and historical, from any source.

Noa has been at SentinelOne for two years and has recently relocated to London. Her teammates are located in the US, Israel and Czech Republic.

Working with people all across the globe provides diverse perspectives,” said Noa. “That diversity allows us to better serve our customers – and each other.

Noa describes the Sentinel team culture as one of relentless innovation.


“Working with people who are excited about our product is the foundation of our culture,” said Noa. “We are working on a vision and executing as a team. The commitment to innovation is something I feel on a daily basis.”

Noa works in product management in partnership with development, marketing and sales to create product roadmaps and execute our mission.

“Data is the key in the world of XDR,” said Noa. “It is difficult for organizations to make sense of such massive amounts of data and this is what we are here to solve. I am proud to collaborate with other leaders in our space to create a vendor-agnostic taxonomy to help all security teams realize better, faster data ingestion and analysis.”

Noa is also excited about our approach to customers as the first priority of the business.

I don’t look at our customers as customers, I look at them as people,” said Noa. “I strive to understand their pain points to deliver what answers their needs and deliver on their requests.

Prior to joining SentinelOne, Noa worked with two other security companies and for the Israel Defense Forces in the Joint Cyber Command. She is grateful for the transparency and equity she feels in her day-to-day interactions at all levels of the organization.

“The people at SentineOne are the best in the business,” said Noa. “It doesn’t matter what your level is at SentinelOne, you are always encouraged to suggest new things. All input is welcomed and encouraged.”

Is SentinelOne a Good Place to Work?

In addition to this Fortune ranking, SentinelOne has received a number of other recent accolades highlighting our best-in-class culture, including:

  • INC. Best Workplaces 2022
  • Fortune Best Workplaces in the Bay Area 2022
  • Fortune Best Workplaces for Millennials 2022
  • Best Workplaces in the Netherlands 2022
  • Best Workplaces in the UK 2022
  • Best Workplaces for Wellbeing in the UK 2022
  • The Bay Area’s Best Places To Work 2022
  • Comparably Best Company Outlook 2022
  • Comparably Best Company For Global Culture 2022
  • Comparably Best Company in the Bay Area 2022
  • Comparably Best Company for Career Growth 2022
  • Comparably Best CEOs for Women 2022
  • Comparably Best CEOs for Diversity 2022
  • Comparably Best Sales Team 2022
  • Comparably Best Engineering Team 2022

To learn more about our award-winning culture and job opportunities, visit our careers page.

The Good, the Bad and the Ugly in Cybersecurity – Week 37

The Good

This week, the DoJ, FBI, and Portuguese authorities dismantled WT1SHOP, a prolific cybercrime marketplace known for the sale of over 5.85 million records of personal identifiable information (PII). One of the largest of its kind, the market sold pilfered login credentials for retailers and financial organizations, email and PayPal accounts, as well as identification cards and network credentials.

Across WT1SHOP’s website and four of its domains, a DoJ report noted approximately 106,273 users and 94 sellers operated in the marketplace as of December 2021. The report also alleges that Moldovian national, Nicolai Colesnicov, was the operator and administrator of WT1SHOP. If convicted, Colesnicov faces up to 10 years in federal prison on the charges of conspiracy and unauthorized device tracking.

With WT1SHOP offline, law enforcement teams around the world add another cyber takedown to their books. This year has seen a number successful darknet seizures including Hydra Market – a notorious, long-running black market for drugs offering cryptocurrency mixing and laundering services, RaidForums – a popular cybercrime marketplace for selling high-profile hacked data, and SSNDOB – a series of websites harboring the social security numbers, names and birthdays of approximately 24 million U.S. citizens.

The continued crackdowns on these marketplaces results in a snowball effect – with each successful bust, investigators find additional leads and data on the next target, eventually making strides in disrupting the greater cybercrime infrastructure and economy.

The Bad

While back-to-school garners mixed emotions across students (some parts excitement, some parts dread) and parents (relief), cybercriminals are, unfortunately, feeling opportunistic. CISA, the FBI, and the Multi-State Information Sharing and Analysis Center (MS-ISAC) released a joint cybersecurity advisory earlier this week against the continuing rise of ransomware attacks on the education sector. The advisory named K-12 institutions as particularly attractive targets to threat actors as their systems are seen as a deep trove of sensitive student data.

The FBI, CISA, and the MS-ISAC expect the number of attacks to increase throughout the school year. Just hours prior to the advisory, Los Angeles Unified (LAUSD) disclosed a ransomware attack on its IT systems. LAUSD represents the second largest school district in the U.S. and supports more 640,000 students.

The attack was reported to have affected the district’s server infrastructure, but instruction, transportation, student meals, and safety systems were not interrupted. However, ransomware actors will typically exfiltrate files from the targeted environment allowing them to ransom their victims later on.

Over the years, the impact of ransomware attacks on schools has ranged from restricted access to critical networks and data to the theft of PII leading to identity crime and extortion. With educators continuing to digitize their administrative assets, protecting sensitive data will be a continuing challenge that requires a coordinated effort across federal leadership, edu-tech vendors, school boards, managed security service providers (MSSPs), and the students and educators themselves.

The Ugly

Energy providers headquartered in the United States, Canada, and Japan have found themselves in the crosshairs of Lazarus APT Group, a North Korean-linked cybercrime group. Security researchers reported this week on a cyberattack campaign specifically targeting energy companies and speculated that the “main goal of these attacks was likely to establish long-term access into victim networks to conduct espionage operations in support of North Korean government objectives.”

During the six-month long campaign, Lazarus employed the Log4j vulnerability to compromise internet-facing VMWare Horizon servers and then gain entry into the targeted enterprise networks. After gaining their initial foothold, the APT group deployed an HTTP bot called “VSingle” and “YamaBot”, a backdoor, to establish long-term persistence. Further analysis revealed that the group used known malware families as well as a custom remote access trojan now dubbed “MagicRAT”.

Attributed by CISA to the North Korean government, Lazarus Group gained notoriety for a string of high-profile attacks including the Sony hack in 2014 and the WannaCry 2.0 global ransomware attack in 2017. Efforts of the Lazarus hacking group have long supported the DPRK’s espionage-driven cyber objectives. The campaign targeting major global energy providers highlights the group’s capability of coordinating various TTP and using a wide range of existing and new, bespoke malware to achieve their operational goals.

In July, the US government put a $10 million reward up for offer in return for information on DPRK-linked threat groups and their members. Bounties like this are a part of an ongoing campaign by the U.S. State Department in search of threat intelligence, particularly concerning malicious intentions on critical infrastructure and interference with federal elections.

Transacting in Person with Strangers from the Internet

Communities like Craigslist, OfferUp, Facebook Marketplace and others are great for finding low- or no-cost stuff that one can pick up directly from a nearby seller, and for getting rid of useful things that don’t deserve to end up in a landfill. But when dealing with strangers from the Internet, there is always a risk that the person you’ve agreed to meet has other intentions.

Nearly all U.S. states now have designated safe trading stations — mostly at local police departments — which ensure that all transactions are handled in plain view of both the authorities and security cameras.

These safe trading places exist because sometimes in-person transactions from the Internet don’t end well for one or more parties involved. The website Craigslistkillers has catalogued news links for at least 132 murders linked to Craigslist transactions since 2015. Many of these killings involved high-priced items like automobiles and consumer electronics, where the prospective buyer apparently intended all along to kill the owner and steal the item offered for sale. Others were motivated simply by a desire to hurt people.

This is not to say that using Craigslist is uniquely risky or dangerous; I’m sure the vast majority of transactions generated by the site end amicably and without physical violence. And that probably holds true for all of Craigslist’s competitors.

Still, the risk of a deal going badly when one meets total strangers from the Internet is not zero, and so it’s only sensible to take a few simple precautions. For example, choosing to transact at a designated safe place such as a police station dramatically reduces the likelihood that anyone wishing you harm would even show up.

I recently stumbled upon one of these designated exchange places by accident, hence my interest in learning more about them. The one I encountered was at a Virginia county sheriff’s office, and it has two parking spots reserved with a sign that reads, “Internet Purchase & Exchange Location: This Area is Under 24 Hour Video Surveillance” [image above].

According to the list maintained at Safetradestations.com, there are four other such designated locations in Northern Virginia. And it appears most states now have them in at least some major cities. Safeexchangepoint.com also has a searchable index of safe trading locations in the United States and Canada.

Granted, not everyone is going to live close to one of these designated trading stations. Or maybe what you want to buy, sell or trade you’d rather not have recorded in front of police cameras. Either way, here are a few tips on staying safe while transacting in real life with strangers from the Internet (compliments of the aforementioned safe trading websites).

The safest exchange points are easily accessible and in a well-lit, public place where transactions are visible to others nearby. Try to arrange a meeting time that is during daylight hours, and consider bringing a friend along — especially when dealing with high-value items like laptops and smart phones.

Safeexchangepoint.com also advises that police or merchants that host their own exchange locations generally won’t get involved in the details of your transaction unless specified otherwise, and that many police departments (but not all) are willing to check the serial number of an item for sale to make sure it’s not known to be stolen property.

Of course, it’s not always practical or possible to haul that old sofa to the local police department, or a used car that isn’t working. In those situations, safetradestations.com has some decent suggestions:

  • Meet at a police station where you can exchange and photocopy each others’ identification papers, such as a driver’s license. Do NOT carry cash to this location.
  • Photocopy the license or identification paper, or use your phone to photograph it.
  • Email the ID information to a friend, or to someone trusted (not to yourself).
  • If you’re selling at home, or going to someone’s home, never be outnumbered. If you’re at home, make sure you have two or three people there — and tell the person who is coming that you will have others with you.
  • At home or an apartment, NEVER let someone go anywhere unaccompanied. Always make sure they are escorted.
  • Never let more than one group come to your home at one time to buy or sell.
  • Beware of common scams, like checks for an amount higher than the amount of the deal; “cashier’s checks” that are forged and presented when the bank is closed.
  • If you are given a cashier’s check, money order or other equivalent, call the bank — at the number listed online, not a number the buyer gives you — to verify the validity of the check.

Cyber Risks in the Education Sector | Why Cybersecurity Needs to Be Top of the Class

With summer on its last legs, the phrase “back to school” can be heard everywhere. For opportunistic threat actors though, this holds an entirely different meaning than it does for students, educators, and guardians. Data for 2022 in the US suggests that the education sector has seen an increase in monthly cyberattack volume since 2021. In the UK, government statistics indicate that 62% of higher education institutions reported experiencing breaches or attacks at least weekly in the previous 12 months.

On a global level, key statistics for 2022 showed that:

  • In July, the education sector experienced double the number of weekly cyberattacks when compared to other industry averages.
  • Education is the most targeted industry with an average of 2297 cyberattacks against organizations each week in the first half of 2022; a 44% increase compared to the first half of 2021.

Cybersecurity in K-12 and higher educational organizations is complicated by multiple factors, including a large and disparate attack surface, varying degrees of cybersecurity awareness among users, restricted budgets, and the need for strategic oversight at the management level. On top of that, schools sit on a honeypot of valuable personal data belonging to students, staff and even parents that is attractive to threat actors.

In this post, we review the risks facing the education sector and discuss recent policy initiatives and cyber defense solutions to help schools, colleges, and universities better manage the cybersecurity challenges they face.

Student PII and Education Software

As school districts lean further into digitizing their methods of teaching, learning, and managing students’ progress, threat actors are leveraging these tools as springboards to accessing student data. In particular, student tracking software is a direct gateway for actors to obtain students’ personally identifiable information (PII), defined by NIST as any data that can be used to distinguish or trace an individual’s identity.

Student tracking software is used by educators to document and manage day-to-day student data such as absenteeism, learning or developmental challenges, disciplinary action plans, reporting, and more. While this type of software greatly helps educators support students, threat actors are targeting the technology for malicious purposes.

The data breach reported by California-based education technology (ed-tech) vendor, Illuminate Education, is a recent example of this. Illuminate Education, which integrates K-12 technology systems for student instruction, assessment, and data analytics fell victim to a data breach, exposing student PII across two of the US’ largest public school systems, New York City Department of Education and Los Angeles Unified School District (LAUSD) with cyber incidents beginning to crop up in other states as well. This breach is especially concerning as Illuminate Education claims to reach 17 million students across 5200 American school districts.

Cybercriminals Hone in on School Data Stockpiles

Schools and the technology they use to collect and manage information offer an attractive target for threat actors as they represent a gateway to data, and lots of it. Schools hold large amounts of sensitive data not only on their students but also on parents and staff.

Unfortunately, where there is data online, there is cyber risk. Sensitive data such as addresses, birthdays, social security numbers, loan applications, and tuition-related banking information can all be targeted by threat actors, who advertise it for sale on online criminal marketplaces and publicly-accessible forums. Even delicate data like notes about a student’s home life, illnesses, ethnicity, test scores, citizenship or migration status can be leveraged by threat actors in reconnaissance efforts and profiling.

In the short term, the consequences of exposing PII stored by the school could lead to a variety of cyberattacks including data breaches through phishing, ransomware attacks, Distributed-Denial-of-Service (DDoS) attacks, and even hacktivism through “zoom-bombing”, “meeting invasions”, and email spamming.

Identity theft is another cyber risk stemming from compromised student PII. Higher education institutions are particularly likely to report impersonation attacks. However, the problem has also been reported in K-12 schools as well. Earlier this year, one family found that their child’s name was being used to apply for a credit card, car loan, and discounts on their “child’s” electric utility account. The child’s personal information had been part of a data breach at their elementary school just a few months prior. From a threat actor’s perspective, credit checks are rarely conducted on children and teenagers, meaning any fraudulent activity carried out in their names may not be noticed until months and years later.

In terms of long-term consequences, there is the potential for stolen PII to be misused in ways that could affect a future college, loan, or even job application. In the Illuminate Education data breach, reports stated that personal details of both present and former students dating back more than a decade had been compromised.

Attention & Action Taken at the Federal Level

Mounting cyberattacks on education are making their way to policymakers at the federal level. In late 2021, President Biden signed the K-12 Cybersecurity Act calling for the Cybersecurity and Infrastructure Security Agency (CISA) to analyze cyber risks faced by elementary and secondary schools and develop recommendations on how to assist them in facing threats.

In March of this year, President Biden’s State of the Union address called on Congress directly to strengthen privacy protections for children, including the collection of their personal data.

This was followed in June 2022 by the Federal Trade Commission (FTC) issuing a policy statement reiterating the Children’s Online Privacy Protection Act (COPPA) for ed-tech providers. Their August statement outlined the following:

  • Providers must not collect more information than is necessary when accessing a child’s online activity,
  • A child’s personal personal information can only be used for the benefit of their school,
  • A child’s personal personal information cannot be kept longer than needed for the specific purpose of data collection, and
  • Processes must be put in place to maintain the privacy, security, and integrity of the child’s personal information.

#StopRansomware | Vice Society

In their latest joint cybersecurity advisory, CISA, the FBI, and the Multi-State Information Sharing and Analysis Center (MS-ISAC) reported on increasing ransomware attacks on the education sector by intrusion, exfiltration, and extortion hacking group, Vice Society.

US authorities warn that “Vice Society actors [are] disproportionately targeting the education sector with ransomware attacks.” The group is known to deploy common ransomware available on the darknet such as HelloKitty and Zeppelin.

Outlining the scope of impact to include “restricted access to networks and data, delayed exams, canceled school days, and unauthorized access to and theft of personal information regarding students and staff”, attacks by this particular hacking group is just one example amongst the wave of attacks on the sector.

As the school year starts this week across American schools, the FBI, CISA, and MS-ISAC anticipate the number of attacks to increase. At a higher level, this joint advisory is part of an ongoing #StopRansomware effort published to detail observed tactics, techniques, and procedures (TTPs) and indicators of compromise (IoCs) to help network defenders and organizations protect against ransomware.

SentinelOne in Defense of the Education Sector

While more restrictive policies are yet to be enforced on ed-tech vendors, vulnerable school districts are not helpless against cyberattacks. Schools can bolster their cyber defenses against ransomware, DDoS attacks, identify theft, and more by implementing a single, robust security platform to ensure full transparency across their networks and endpoints.

Many schools without the resources to fund and staff dedicated security teams choose to outsource cybersecurity to an MSP (Managed Service Provider) or MSSP (Managed Security Service Provider), an increasing number of which rely on SentinelOne’s Singularity platform to provide managed security services to schools and other organizations.

Educational institutions can partner directly with SentinelOne and take advantage of AI-powered prevention, detection, response, and advanced threat hunting capabilities. SentinelOne’s single autonomous platform delivers protection for Chromebooks, Macs, Windows and Linux devices, as well as server and cloud workload protection. The school’s IT team can see everything happening across their network at machine-speed, staying ahead of threat actors and preventing malicious behavior from developing into full-blown cyberattacks. Schools, colleges, and universities can also rely on SentinelOne’s integrated Identity solution to help prevent breaches through the abuse or theft of credentials.

Conclusion

Protecting the data, services and users within educational organizations is a challenge that requires a coordinated strategy. The complexity of the threat along with the squeeze on resources means planners need to be aware of the risks they face and current government guidelines. Partnering with external cybersecurity providers and deploying a modern, trusted security solution can help address these challenges within the school’s budget.

Leveraging our deep industry knowledge and experience with fighting back privileged escalation and lateral movement, SentinelOne’s Singularity™ XDR delivers comprehensive security with autonomous protection. Through AI-based behavioral detection and superior detection of ransomware attacks, SentinelOne eases the burden on under-resourced school IT teams, leaving staff to focus their attention on operational tasks instead.

To learn more about how SentinelOne helps protect K-12 and higher education, please contact us or request a demo.

Singularity for ChromeOS
Real-Time Protection for Chromebooks

Accelerating Your Cloud Security with Workload Protection

As more organizations make the shift towards hybrid and cloud environments, security teams need a new way to keep their cloud workloads safe from cyber threats. Cloud services offer organizations a scalability that isn’t possible with on-premise infrastructure as well as a boost to efficiency; however, the shift also comes with unique considerations when it comes to security. Outside the scope of your typical cybersecurity practices, cloud computing requires organizations to secure containers, virtual machines, serverless workloads and Kubernetes whether the cloud is public, private, or a hybrid of both.

While the cloud has hugely supported a modern, digital means of collaboration and operation, especially since the COVID-19 pandemic, its adoption also adds cyber risks that are associated with it. Organizations can mitigate these risks by implementing a holistic security strategy focused on workload protection to protect their cloud environments.

Defining a Security Strategy for Cloud

The increase of remote work has given rise to cybersecurity threats to both cloud and hybrid workspaces. With new attack techniques plentiful in the vast threat landscape, threat actors are taking advantage of the larger attack surface as organizations start to store more data and offer services in the cloud.

Defining a security strategy for cloud starts with discovery based on an organizations’ core business objectives, principles, and priorities. No security strategy that is out of alignment with an organization’s goals ever proves to be successful – how can you fully protect what you can’t see? Before beginning the migration into cloud, invest time and effort in mapping out the key aspects of your organization, your attack surface, and their relationship to the cloud security you need.

  • What are my organization’s most critical assets/data?
    • What compliance regulations or requirements does my cloud need to meet in terms of storage?
  • What are the most critical cloud threats my organization faces?
    • What processes and technology does my organization have in place to secure those threats?
  • What are the immediate and long-term impacts should my organization face a successful cyberattack on the cloud?
    • What incident response plans and processes does my organization have in place?
  • What internal and external vulnerabilities does my organization’s cloud have?
    • What is the likelihood of these vulnerabilities being exploited?
    • What processes and technology does my organization have to address these vulnerabilities?

When not managed properly, cloud computing can actually end up exposing organizations to opportunistic cyberattacks. Clouds are particularly vulnerable to misconfiguration, Active Directory vulnerabilities, insider threats, and supply chain attacks. The likelihood of these threat activities targeting the cloud will continue to grow in number, so having a strong cloud security strategy puts preventative measures in place against breach and data loss.

Choosing the Right Cloud Security Technology

Planning, building, and enforcing the organization’s cloud strategy will be a main area of concern for CISOs and security teams. A large part of that strategy will be the direct result of choosing the right security solution for an organization’s cloud setup. The right solution for an organization’s cloud needs to be scalable, easy to manage, and able to defend against increasingly complex cloud-related cyber threats.

These are the key aspects that a cloud security solution must address:

  • Visibility Management – Cloud-based environments are easy to scale up in response to growing data volumes, which makes them a popular solution for organizations wanting to improve their flexibility and agility. As easy as it is to spin up new workloads in the cloud though, lack of visibility and misconfiguration of those workloads could leave them exposed to potential security vulnerabilities. A foundational step is to maintain deep visibility into what is running in your cloud at all times to limit exposure and reduce risk.
  • Integration Compatibility – Larger organizations with established tech stacks must think about tool compatibility and the quality of their integrations. Especially for organizations who have hybrid environments, existing tools must be able to integrate with the cloud. Having seamless integration between your cloud and your security tools ensures nothing operates in isolation and that data is synchronized in a reliable exchange.
  • Real-time Detection – With enough time and resources, threat actors frequently meet their goals. This makes fast detection the keystone in preventing actors from inflicting critical damage to your cloud environment. With the time between initial intrusion and lateral movement getting shorter, quick detection time is a crucial element of an organization’s defenses.
  • Autonomous Response – A solution that employs artificial intelligence (AI) and machine learning (ML) can be leveraged very effectively against modern threat actors from attacking your cloud. AI technology augments security teams by automating the interpretation of attack signals, prioritizing alerts and incidents, and adapting responses based on the scale and speed of the attacker.
  • Data Compliance – Cybersecurity and compliance go hand in hand. Cloud security technology should help organizations meet the requirements of the regulation frameworks they abide by and allow them to use, store, manage, transmit, and protect sensitive data in accordance with applicable controls. This includes, but is not limited to, data encryption and a robust endpoint protection (EPP) solution.

Singularity Cloud | SentinelOne’s Approach to Securing the Cloud

SentinelOne enables organizations to protect their endpoints across all cloud environments, public, private, and hybrid, through Singularity Cloud. With thousands of accounts spread across multiple clouds, organizations need the right security in place for their cloud infrastructure. Singularity Cloud works by extending distributed, autonomous endpoint protection, detection, and response to compute workloads running in both public and private clouds, as well as on-prem data centers.

Within the current cyber landscape, cloud workload protection platforms (CWPP) are the final line of defense in a multi-layer cloud security strategy. Organizations rely on CWPP’s like Singularity Cloud for autonomous, real-time detection as well as remediation of complex threats at the VM and K8 pod level with no need for human detection. Further, Singularity Cloud’s runtime protection of containerized workloads identifies and kills unauthorized processes such as malware, ransomware, and more.

  • AI-Powered Cloud Workload Protection – Behavioral AI detects unknown threats such as zero-day exploits and indicators of compromise consistent with novel ransomware and then quarantines them in real-time. Singularity Cloud protects runtime containers without container interference for Linux, Windows servers, and VMs.
  • Enterprise-Grade EPP & EDR – Get full endpoint detection and response as well as container coverage in one SentinelOne agent. Singularity Cloud allows for complete container visibility with one agent per node and without pod instrumentation.
  • Enterprise Management & Deployment – Choose to auto-deploy Kubernetes Sentinel Agent, a component of Singularity Cloud to EKS, AKS, and GKE clusters, or Linux and Windows Server Sentinel Agents to AWS EC2, Azure VM, and Google Compute Engine.

Conclusion

Opportunistic threat actors attacking clouds count on the fact that cloud networks are large, complex, and require in-depth configuration and management. This means it is critical for organizations to choose the right cloud security platform in support of their overarching security strategy. SentinelOne is here to help you improve your cloud security plan and fuse autonomous threat hunting, EDR capability, and security together to fit your business. Contact us today or book a demo to see how Singularity Cloud brings agility, AI-powered security, and compliance to organizations globally.

Singularity Cloud
Simplifying security of cloud VMs and containers, no matter their location, for maximum agility, security, and compliance.

Violence-as-a-Service: Brickings, Firebombings & Shootings for Hire

A 21-year-old New Jersey man has been arrested and charged with stalking in connection with a federal investigation into groups of cybercriminals who are settling scores by hiring people to carry out physical attacks on their rivals. Prosecutors say the defendant recently participated in several of these schemes — including firing a handgun into a Pennsylvania home and torching a residence in another part of the state with a Molotov Cocktail.

Patrick McGovern-Allen of Egg Harbor Township, N.J. was arrested on Aug. 12 on a warrant from the U.S. Federal Bureau of Investigation. An FBI complaint alleges McGovern-Allen was part of a group of co-conspirators who are at the forefront of a dangerous escalation in coercion and intimidation tactics increasingly used by competing cybercriminal groups.

Prosecutors say that around 2 a.m. on Jan 2, 2022, McGovern-Allen and an unidentified co-conspirator fired multiple handgun rounds into a residence in West Chester, Pa. Fortunately, none of the residents inside the home at the time were injured. But prosecutors say the assailants actually recorded video of the attack as “proof” that the shooting had been carried out.

A copy of that video was obtained by KrebsOnSecurity. According to investigators, McGovern-Allen was one of the shooters, who yelled “Justin Active was here” as they haphazardly fired at least eight rounds into the lower story of the West Chester residence.

On Dec. 18, 2021, police in Abington Township, Pa., responded to reports of a house fire from homeowners who said it sounded like something was thrown at their residence just prior to the fire.

Weeks later, on the day of the shooting in West Chester, a detective with the Westtown East Goshen Police Department contacted the Abington police and shared another video that was circulating on several online message boards that appeared to show two individuals setting fire to the Abington Township residence. The criminal complaint said the two police officers agreed the same suspect was present in both videos.

A copy of that video also was obtained by KrebsOnSecurity, and it shows at least two individuals smashing a window, then lighting a rag-soaked Mad Dog 20/20 grape wine bottle and hurling it at the side of the home [Update: My apologies for the file download link, but YouTube just deleted both of the videos included in this story — for allegedly violating their community standards].

“The Molotov cocktail caused the immediate surrounding area to ignite, including the siding of the house, grass, and the wooden chair,” the government’s complaint against McGovern-Allen states. “The two suspects then fled on foot toward the street and begin yelling something when the video stops.”

The government mentions the victims only by their initials — “K.M.” in the shooting and “A.R.” in the firebombing — but said both had been the target of previous harassment by rival cybercriminal groups that included swatting attacks, wherein the perpetrators spoof a distress call to the police about a hostage situation, suicide or bomb threat with the goal of sending a heavily-armed police response to a targeted address.

A number of previous swatting incidents have turned deadly. But these more “hands-on” and first person attacks are becoming increasingly common within certain cybercriminal communities, particularly those engaged in SIM swapping, a crime in which identity thieves hijack a target’s mobile phone number and use that to wrest control over the victim’s various online accounts and identities.

The complaint mentions a handle and user ID allegedly used by McGovern-Allen’s online persona “Tongue” on the Discord chat service, (user: “Tongue#0001”).

“In the chats, [Tongue] tells other Discord users that he was the person who shot K.M.’s house and that he was willing to commit firebombings using Molotov Cocktails,” the complaint alleges. “For example, in one Discord chat from March 2022, [the defendant] states ‘if you need anything done for $ lmk [“let me know”]/I did a shooting/Molotov/but I can also do things for ur entertainment.”

KrebsOnsecurity reviewed hundreds of chat records tied to this Tongue alias, and it appears both attacks were motivated by a desire to get back at a rival cybercriminal by attacking the female friends of that rival.

Recall that the shooters in the West Chester, Pa. incident shouted “Justin Active was here.” Justin Active is the nickname of an individual who is just as active in the same cybercriminal channels, but who has vehemently denied knowledge of or participation in the shooting. Justin Active said on Telegram that the person targeted in the shooting was his ex-girlfriend, and that the firebombing targeted another friend of his.

Justin Active has claimed for months that McGovern-Allen was responsible for both attacks, saying they were intended as an intimidation tactic against him. “DO THE PATRICK MCGOVERN ALLEN RAID DANCE!,” Justin Active’s alias “Nutcase68” shouted on Telegram on Aug. 12, the same day McGovern-Allen was arrested by authorities.

Justin Active’s version of events seems to be supported by a reference in the criminal complaint to an April 2, 2022 chat in which Tongue explained the reason for the shooting.

“The video/is [K]’s house/getting shit/shot/justin active/ was her current bf/ the reason it happened,” Tongue explained. “So that’s why Justin active was there.”

The Telegram chat channels that Justin Active and Tongue both frequented have hundreds to thousands of members each, and some of the more interesting solicitations on these communities are job offers for in-person assignments and tasks that can be found if one searches for posts titled, “If you live near,” or “IRL job” — short for “in real life” job.

A number of these classified ads are in service of performing “brickings,” where someone is hired to visit a specific address and toss a brick through the target’s window.

“If you live near Edmonton Canada dm me need someone bricked,” reads on Telegram message on May 31, 2022.

“If you live near [address redacted] Lakewood, CA, dm [redacted] Paying 3k to slash the tires,” reads another help wanted ad in the same channel on Feb. 24, 2022. “If you live near here and can brick them, dm [address omitted] Richland, WA,” reads another from that same day.

McGovern-Allen was in the news not long ago. According to a Sept. 2020 story from The Press of Atlantic City, a then 19-year-old Patrick McGovern Allen was injured after driving into a building and forcing residents from their home.

“Police found a 2007 Lexus, driven by Patrick McGovern-Allen, 19, that had lost control and left the road, crashing into the eastern end of the 1600 building,” the story recounted. “The car was driven through the steps that provide access to the second-floor apartments, destroying them, and also caused damage to the outer wall.”

A search on the Inmate Locator of the U.S. Bureau of Prisons website shows that McGovern-Allen remains in federal custody at a detention facility in Philadelphia. He’s currently represented by a public defender who has not responded to requests for comment.

A copy of the criminal complaint against McGovern-Allen is available here (PDF).

ANALYSIS

Many of the individuals involved in paying others to commit these physical attacks are also frequent participants in several Telegram channels focused singularly on SIM swapping activity. As a result, the vast majority of the people being targeted for brickings and other real-life physical assaults tend to be other cybercriminals involved in SIM swapping crimes (or individuals on the periphery of that scene).

There are dozens of SIM swappers who are now teenage or 20-something millionaires, by virtue of having stolen vast sums of cryptocurrencies from SIM swapping victims. And now many of these same individuals are finding that communities like Telegram can be leveraged to hire physical harassment and intimidation of their rivals and competitors.

The primary barrier to hiring someone to brick a home or slash some tires seems to be the costs involved: A number of solicitations for these services advertised payment of $3,000 or more upon proof of successful completion, which usually involves recording the attack and hiring a getaway driver in the town where the crime is to take place (calling a cab or hailing an Uber from the scene of a bricking isn’t the brightest idea).

My fear is these violence-as-a-service offerings will at some point migrate outside of the SIM swapping communities. This is precisely what happened with swatting, which for years was a crime perpetrated almost exclusively against online gamers and people streaming their games online. These days, swatting attacks are commonly used by SIM swapping groups as a way to harass and extort regular Internet users into giving up prized social media account names that can be resold for thousands of dollars.