Techniques

Adversaries may leverage traffic mirroring in order to automate data exfiltration over compromised infrastructure. Traffic mirroring is a native feature for some devices, often used for network analysis. For example, devices may be configured to forward network traffic to one or more destinations for analysis by a network analyzer or other monitoring device. [1][2]

Adversaries may abuse traffic mirroring to mirror or redirect network traffic through other infrastructure they control. Malicious modifications to network devices to enable traffic redirection may be possible through ROMMONkit or Patch System Image.[3][4]

Many cloud-based environments also support traffic mirroring. For example, AWS Traffic Mirroring, GCP Packet Mirroring, and Azure vTap allow users to define specified instances to collect traffic from and specified targets to send collected traffic to.[5][6][7]

Adversaries may use traffic duplication in conjunction with Network Sniffing, Input Capture, or Adversary-in-the-Middle depending on the goals and objectives of the adversary.

Adversaries may use traffic signaling to hide open ports or other malicious functionality used for persistence or command and control. Traffic signaling involves the use of a magic value or sequence that must be sent to a system to trigger a special response, such as opening a closed port or executing a malicious task. This may take the form of sending a series of packets with certain characteristics before a port will be opened that the adversary can use for command and control. Usually this series of packets consists of attempted connections to a predefined sequence of closed ports (i.e. Port Knocking), but can involve unusual flags, specific strings, or other unique characteristics. After the sequence is completed, opening a port may be accomplished by the host-based firewall, but could also be implemented by custom software.

Adversaries may also communicate with an already open port, but the service listening on that port will only respond to commands or trigger other malicious functionality if passed the appropriate magic value(s).

The observation of the signal packets to trigger the communication can be conducted through different methods. One means, originally implemented by Cd00r [1], is to use the libpcap libraries to sniff for the packets in question. Another method leverages raw sockets, which enables the malware to use ports that are already open for use by other programs.

On network devices, adversaries may use crafted packets to enable Network Device Authentication for standard services offered by the device such as telnet. Such signaling may also be used to open a closed service port such as telnet, or to trigger module modification of malware implants on the device, adding, removing, or changing malicious capabilities. Adversaries may use crafted packets to attempt to connect to one or more (open or closed) ports, but may also attempt to connect to a router interface, broadcast, and network address IP on the same port in order to achieve their goals and objectives.[2][3][4] To enable this traffic signaling on embedded devices, adversaries must first achieve and leverage Patch System Image due to the monolithic nature of the architecture.

Adversaries may also use the Wake-on-LAN feature to turn on powered off systems. Wake-on-LAN is a hardware feature that allows a powered down system to be powered on, or woken up, by sending a magic packet to it. Once the system is powered on, it may become a target for lateral movement.[5][6]

Adversaries may exfiltrate data by transferring the data, including through sharing/syncing and creating backups of cloud environments, to another cloud account they control on the same service.

A defender who is monitoring for large transfers to outside the cloud environment through normal file transfers or over command and control channels may not be watching for data transfers to another account within the same cloud provider. Such transfers may utilize existing cloud provider APIs and the internal address space of the cloud provider to blend into normal traffic or avoid data transfers over external network interfaces.[1]

Adversaries may also use cloud-native mechanisms to share victim data with adversary-controlled cloud accounts, such as creating anonymous file sharing links or, in Azure, a shared access signature (SAS) URI.[2]

Incidents have been observed where adversaries have created backups of cloud instances and transferred them to separate accounts.[3]

Adversaries may target devices that are transient across ICS networks and external networks. Normally, transient assets are brought into an environment by authorized personnel and do not remain in that environment on a permanent basis. [1] Transient assets are commonly needed to support management functions and may be more common in systems where a remotely managed asset is not feasible, external connections for remote access do not exist, or 3rd party contractor/vendor access is required.

Adversaries may take advantage of transient assets in different ways. For instance, adversaries may target a transient asset when it is connected to an external network and then leverage its trusted access in another environment to launch an attack. They may also take advantage of installed applications and libraries that are used by legitimate end-users to interact with control system devices.

Transient assets, in some cases, may not be deployed with a secure configuration leading to weaknesses that could allow an adversary to propagate malicious executable code, e.g., the transient asset may be infected by malware and when connected to an ICS environment the malware propagates onto other systems.

Adversaries may alter data en route to storage or other systems in order to manipulate external outcomes or hide activity.[1][2] By manipulating transmitted data, adversaries may attempt to affect a business process, organizational understanding, and decision making.

Manipulation may be possible over a network connection or between system processes where there is an opportunity deploy a tool that will intercept and change information. The type of modification and the impact it will have depends on the target transmission mechanism as well as the goals and objectives of the adversary. For complex systems, an adversary would likely need special expertise and possibly access to specialized software related to the system that would typically be gained through a prolonged information gathering campaign in order to have the desired impact.

Adversaries may alter data en route to storage or other systems in order to manipulate external outcomes or hide activity, thus threatening the integrity of the data.[1][2] By manipulating transmitted data, adversaries may attempt to affect a business process, organizational understanding, and decision making.

Manipulation may be possible over a network connection or between system processes where there is an opportunity deploy a tool that will intercept and change information. The type of modification and the impact it will have depends on the target transmission mechanism as well as the goals and objectives of the adversary. For complex systems, an adversary would likely need special expertise and possibly access to specialized software related to the system that would typically be gained through a prolonged information gathering campaign in order to have the desired impact.

Adversaries may alter data en route to storage or other systems in order to manipulate external outcomes or hide activity. By manipulating transmitted data, adversaries may attempt to affect a business process, organizational understanding, or decision making.

Manipulation may be possible over a network connection or between system processes where there is an opportunity to deploy a tool that will intercept and change information. The type of modification and the impact it will have depends on the target transmission mechanism as well as the goals and objectives of the adversary. For complex systems, an adversary would likely need special expertise and possibly access to specialized software related to the system, typically gained through a prolonged information gathering campaign, in order to have the desired impact.

One method to achieve Transmitted Data Manipulation is by modifying the contents of the device clipboard. Malicious applications may monitor clipboard activity through the `ClipboardManager.OnPrimaryClipChangedListener` interface on Android to determine when clipboard contents have changed. Listening to clipboard activity, reading clipboard contents, and modifying clipboard contents requires no explicit application permissions and can be performed by applications running in the background. However, this behavior has changed with the release of Android 10.

Adversaries may use Transmitted Data Manipulation to replace text prior to being pasted. For example, replacing a copied Bitcoin wallet address with a wallet address that is under adversarial control.

Transmitted Data Manipulation was seen within the Android/Clipper.C trojan. This sample was detected by ESET in an application distributed through the Google Play Store targeting cryptocurrency wallet numbers.[1]

Adversaries may abuse Microsoft transport agents to establish persistent access to systems. Microsoft Exchange transport agents can operate on email messages passing through the transport pipeline to perform various tasks such as filtering spam, filtering malicious attachments, journaling, or adding a corporate signature to the end of all outgoing emails.[1][2] Transport agents can be written by application developers and then compiled to .NET assemblies that are subsequently registered with the Exchange server. Transport agents will be invoked during a specified stage of email processing and carry out developer defined tasks.

Adversaries may register a malicious transport agent to provide a persistence mechanism in Exchange Server that can be triggered by adversary-specified email events.[2] Though a malicious transport agent may be invoked for all emails passing through the Exchange transport pipeline, the agent can be configured to only carry out specific tasks in response to adversary defined criteria. For example, the transport agent may only carry out an action like copying in-transit attachments and saving them for later exfiltration if the recipient email address matches an entry on a list provided by the adversary.

Adversaries may establish persistence by executing malicious content triggered by an interrupt signal. The trap command allows programs and shells to specify commands that will be executed upon receiving interrupt signals. A common situation is a script allowing for graceful termination and handling of common keyboard interrupts like ctrl+c and ctrl+d.

Adversaries can use this to register code to be executed when the shell encounters specific interrupts as a persistence mechanism. Trap commands are of the following format trap 'command list' signals where "command list" will be executed when "signals" are received.[1][2]

The trap command allows programs and shells to specify commands that will be executed upon receiving interrupt signals. A common situation is a script allowing for graceful termination and handling of common keyboard interrupts like ctrl+c and ctrl+d. Adversaries can use this to register code to be executed when the shell encounters specific interrupts either to gain execution or as a persistence mechanism. Trap commands are of the following format trap 'command list' signals where "command list" will be executed when "signals" are received.[1][2]

Adversaries may add new domain trusts, modify the properties of existing domain trusts, or otherwise change the configuration of trust relationships between domains and tenants to evade defenses and/or elevate privileges.Trust details, such as whether or not user identities are federated, allow authentication and authorization properties to apply between domains or tenants for the purpose of accessing shared resources.[1] These trust objects may include accounts, credentials, and other authentication material applied to servers, tokens, and domains.

Manipulating these trusts may allow an adversary to escalate privileges and/or evade defenses by modifying settings to add objects which they control. For example, in Microsoft Active Directory (AD) environments, this may be used to forge SAML Tokens without the need to compromise the signing certificate to forge new credentials. Instead, an adversary can manipulate domain trusts to add their own signing certificate. An adversary may also convert an AD domain to a federated domain using Active Directory Federation Services (AD FS), which may enable malicious trust modifications such as altering the claim issuance rules to log in any valid set of credentials as a specified user.[2]

An adversary may also add a new federated identity provider to an identity tenant such as Okta or AWS IAM Identity Center, which may enable the adversary to authenticate as any user of the tenant.[3] This may enable the threat actor to gain broad access into a variety of cloud-based services that leverage the identity tenant. For example, in AWS environments, an adversary that creates a new identity provider for an AWS Organization will be able to federate into all of the AWS Organization member accounts without creating identities for each of the member accounts.[4]

Adversaries may take advantage of trusted developer utilities to proxy execution of malicious payloads. There are many utilities used for software development related tasks that can be used to execute code in various forms to assist in development, debugging, and reverse engineering.[1][2][3][4] These utilities may often be signed with legitimate certificates that allow them to execute on a system and proxy execution of malicious code through a trusted process that effectively bypasses application control solutions.

Smart App Control is a feature of Windows that blocks applications it considers potentially malicious from running by verifying unsigned applications against a known safe list from a Microsoft cloud service before executing them.[5] However, adversaries may leverage "reputation hijacking" to abuse an operating system’s trust of safe, signed applications that support the execution of arbitrary code. By leveraging Trusted Developer Utilities Proxy Execution to run their malicious code, adversaries may bypass Smart App Control protections.[6]

Adversaries may breach or otherwise leverage organizations who have access to intended victims. Access through trusted third party relationship abuses an existing connection that may not be protected or receives less scrutiny than standard mechanisms of gaining access to a network.

Organizations often grant elevated access to second or third-party external providers in order to allow them to manage internal systems as well as cloud-based environments. Some examples of these relationships include IT services contractors, managed security providers, infrastructure contractors (e.g. HVAC, elevators, physical security). The third-party provider's access may be intended to be limited to the infrastructure being maintained, but may exist on the same network as the rest of the enterprise. As such, Valid Accounts used by the other party for access to internal network systems may be compromised and used.[1]

In Office 365 environments, organizations may grant Microsoft partners or resellers delegated administrator permissions. By compromising a partner or reseller account, an adversary may be able to leverage existing delegated administrator relationships or send new delegated administrator offers to clients in order to gain administrative control over the victim tenant.[2]

Adversaries may register Uniform Resource Identifiers (URIs) to intercept sensitive data.

Applications regularly register URIs with the operating system to act as a response handler for various actions, such as logging into an app using an external account via single sign-on. This allows redirections to that specific URI to be intercepted by the application. If a malicious application were to register for a URI that was already in use by a genuine application, the malicious application may be able to intercept data intended for the genuine application or perform a phishing attack against the genuine application. Intercepted data may include OAuth authorization codes or tokens that could be used by the malicious application to gain access to resources.[1][2]

Adversaries may register Uniform Resource Identifiers (URIs) to intercept sensitive data.

Applications regularly register URIs with the operating system to act as a response handler for various actions, such as logging into an app using an external account via single sign-on. This allows redirections to that specific URI to be intercepted by the application. If an adversary were to register for a URI that was already in use by a genuine application, the adversary may be able to intercept data intended for the genuine application or perform a phishing attack against the genuine application. Intercepted data may include OAuth authorization codes or tokens that could be used by the adversary to gain access to protected resources.[1][2]

An iOS application may be able to maliciously claim a URL scheme, allowing it to intercept calls that are meant for a different application[1][2]. This technique, for example, could be used to capture OAuth authorization codes[3] or to phish user credentials[4].

Adversaries may maintain persistence through executing malicious content triggered using udev rules. Udev is the Linux kernel device manager that dynamically manages device nodes, handles access to pseudo-device files in the `/dev` directory, and responds to hardware events, such as when external devices like hard drives or keyboards are plugged in or removed. Udev uses rule files with `match keys` to specify the conditions a hardware event must meet and `action keys` to define the actions that should follow. Root permissions are required to create, modify, or delete rule files located in `/etc/udev/rules.d/`, `/run/udev/rules.d/`, `/usr/lib/udev/rules.d/`, `/usr/local/lib/udev/rules.d/`, and `/lib/udev/rules.d/`. Rule priority is determined by both directory and by the digit prefix in the rule filename.[1][2]

Adversaries may abuse the udev subsystem by adding or modifying rules in udev rule files to execute malicious content. For example, an adversary may configure a rule to execute their binary each time the pseudo-device file, such as `/dev/random`, is accessed by an application. Although udev is limited to running short tasks and is restricted by systemd-udevd's sandbox (blocking network and filesystem access), attackers may use scripting commands under the action key `RUN+=` to detach and run the malicious content’s process in the background to bypass these controls.[3]

Adversaries may send unauthorized command messages to instruct control system assets to perform actions outside of their intended functionality, or without the logical preconditions to trigger their expected function. Command messages are used in ICS networks to give direct instructions to control systems devices. If an adversary can send an unauthorized command message to a control system, then it can instruct the control systems device to perform an action outside the normal bounds of the device's actions. An adversary could potentially instruct a control systems device to perform an action that will cause an Impact. [1]

In the Dallas Siren incident, adversaries were able to send command messages to activate tornado alarm systems across the city without an impending tornado or other disaster. [2] [3]

Adversaries may send unauthorized messages to ICS systems and devices to evade defenses or manipulate processes. Unauthorized messages can be categorized as either reporting messages that contain telemetry data about the current state of systems, devices, and processes or as command messages which instruct systems and devices on how to operate. By injecting unauthorized messages, adversaries can make it appear as if everything is working correctly when it isn’t, trigger alarms to misdirect personnel or impact processes, and manipulate controls to disrupt processes.[1]

Adversaries may send unauthorized messages in an ICS environment using software found within the environment (living-off-the-land, vendor-specific interfaces, etc.), custom tooling leveraging OT protocols and libraries, or by positioning themselves between systems and devices and injecting messages into the communications such as the case with an Adversary-in-the-Middle attack.

Adversaries may conduct C2 communications over a non-standard port to bypass proxies and firewalls that have been improperly configured.

Adversaries may include functionality in malware that uninstalls the malicious application from the device. This can be achieved by: * Abusing device owner permissions to perform silent uninstallation using device owner API calls. * Abusing root permissions to delete files from the filesystem. * Abusing the accessibility service. This requires sending an intent to the system to request uninstallation, and then abusing the accessibility service to click the proper places on the screen to confirm uninstallation.

Adversaries may include functionality in malware that uninstalls the malicious application from the device. This can be achieved by:

* Abusing device owner permissions to perform silent uninstallation using device owner API calls. * Abusing root permissions to delete files from the filesystem. * Abusing the accessibility service. This requires an intent be sent to the system to request uninstallation, and then abusing the accessibility service to click the proper places on the screen to confirm uninstallation.

Adversaries may abuse Unix shell commands and scripts for execution. Unix shells are the primary command prompt on Linux, macOS, and ESXi systems, though many variations of the Unix shell exist (e.g. sh, ash, bash, zsh, etc.) depending on the specific OS or distribution.[1][2] Unix shells can control every aspect of a system, with certain commands requiring elevated privileges.

Unix shells also support scripts that enable sequential execution of commands as well as other typical programming operations such as conditionals and loops. Common uses of shell scripts include long or repetitive tasks, or the need to run the same set of commands on multiple systems.

Adversaries may abuse Unix shells to execute various commands or payloads. Interactive shells may be accessed through command and control channels or during lateral movement such as with SSH. Adversaries may also leverage shell scripts to deliver and execute multiple commands on victims or as part of payloads used for persistence.

Some systems, such as embedded devices, lightweight Linux distributions, and ESXi servers, may leverage stripped-down Unix shells via Busybox, a small executable that contains a variety of tools, including a simple shell.

Adversaries may abuse Unix shell commands and scripts for execution. Unix shells are the underlying command prompts on Android and iOS devices. Unix shells can control every aspect of a system, with certain commands requiring elevated privileges that are only accessible if the device has been rooted or jailbroken.

Unix shells also support scripts that enable sequential execution of commands as well as other typical programming operations such as conditionals and loops. Common uses of shell scripts include long or repetitive tasks, or the need to run the same set of commands on multiple systems.

Adversaries may abuse Unix shells to execute various commands or payloads. Interactive shells may be accessed through command and control channels or during lateral movement such as with SSH. Adversaries may also leverage shell scripts to deliver and execute multiple commands on victims or as part of payloads used for persistence.

If the device has been rooted or jailbroken, adversaries may locate and invoke a superuser binary to elevate their privileges and interact with the system as the root user. This dangerous level of permissions allows the adversary to run special commands and modify protected system files.

Adversaries may establish persistence through executing malicious commands triggered by a user’s shell. User Unix Shells execute several configuration scripts at different points throughout the session based on events. For example, when a user opens a command-line interface or remotely logs in (such as via SSH) a login shell is initiated. The login shell executes scripts from the system (/etc) and the user’s home directory (~/) to configure the environment. All login shells on a system use /etc/profile when initiated. These configuration scripts run at the permission level of their directory and are often used to set environment variables, create aliases, and customize the user’s environment. When the shell exits or terminates, additional shell scripts are executed to ensure the shell exits appropriately.

Adversaries may attempt to establish persistence by inserting commands into scripts automatically executed by shells. Using bash as an example, the default shell for most GNU/Linux systems, adversaries may add commands that launch malicious binaries into the /etc/profile and /etc/profile.d files.[1][2] These files typically require root permissions to modify and are executed each time any shell on a system launches. For user level permissions, adversaries can insert malicious commands into ~/.bash_profile, ~/.bash_login, or ~/.profile which are sourced when a user opens a command-line interface or connects remotely.[3][4] Since the system only executes the first existing file in the listed order, adversaries have used ~/.bash_profile to ensure execution. Adversaries have also leveraged the ~/.bashrc file which is additionally executed if the connection is established remotely or an additional interactive shell is opened, such as a new tab in the command-line interface.[5][3][6][7] Some malware targets the termination of a program to trigger execution, adversaries can use the ~/.bash_logout file to execute malicious commands at the end of a session.

For macOS, the functionality of this technique is similar but may leverage zsh, the default shell for macOS 10.15+. When the Terminal.app is opened, the application launches a zsh login shell and a zsh interactive shell. The login shell configures the system environment using /etc/profile, /etc/zshenv, /etc/zprofile, and /etc/zlogin.[8][9][10][11] The login shell then configures the user environment with ~/.zprofile and ~/.zlogin. The interactive shell uses the ~/.zshrc to configure the user environment. Upon exiting, /etc/zlogout and ~/.zlogout are executed. For legacy programs, macOS executes /etc/bashrc on startup.