Techniques

Adversaries may break out of a container or virtualized environment to gain access to the underlying host. This can allow an adversary access to other containerized or virtualized resources from the host level or to the host itself. In principle, containerized / virtualized resources should provide a clear separation of application functionality and be isolated from the host environment.[1]

There are multiple ways an adversary may escape from a container to a host environment. Examples include creating a container configured to mount the host’s filesystem using the bind parameter, which allows the adversary to drop payloads and execute control utilities such as cron on the host; utilizing a privileged container to run commands or load a malicious kernel module on the underlying host; or abusing system calls such as `unshare` and `keyctl` to escalate privileges and steal secrets.[2][3][4][5][6][7]

Additionally, an adversary may be able to exploit a compromised container with a mounted container management socket, such as `docker.sock`, to break out of the container via a Container Administration Command.[5] Adversaries may also escape via Exploitation for Privilege Escalation, such as exploiting vulnerabilities in global symbolic links in order to access the root directory of a host machine.[8]

In ESXi environments, an adversary may exploit a vulnerability in order to escape from a virtual machine into the hypervisor.[9]

Gaining access to the host may provide the adversary with the opportunity to achieve follow-on objectives, such as establishing persistence, moving laterally within the environment, accessing other containers or virtual machines running on the host, or setting up a command and control channel on the host.

Adversaries may create and cultivate accounts with services that can be used during targeting. Adversaries can create accounts that can be used to build a persona to further operations. Persona development consists of the development of public information, presence, history and appropriate affiliations. This development could be applied to social media, website, or other publicly available information that could be referenced and scrutinized for legitimacy over the course of an operation using that persona or identity.[1][2]

For operations incorporating social engineering, the utilization of an online persona may be important. These personas may be fictitious or impersonate real people. The persona may exist on a single site or across multiple sites (ex: Facebook, LinkedIn, Twitter, Google, GitHub, Docker Hub, etc.). Establishing a persona may require development of additional documentation to make them seem real. This could include filling out profile information, developing social networks, or incorporating photos.[1][2]

Establishing accounts can also include the creation of accounts with email providers, which may be directly leveraged for Phishing for Information or Phishing.[3] In addition, establishing accounts may allow adversaries to abuse free services, such as registering for trial periods to Acquire Infrastructure for malicious purposes.[4]

Adversaries may block access to Ethernet communications to prevent instructions or configurations messages from reaching target systems and devices. Ethernet connections allow for communications between IT and OT systems and devices. Blocking Ethernet communications may also block command and reporting messages.[1]

An adversary may block Ethernet communications by disabling network interfaces, Service Stop, or conducting an Adversary-in-the-Middle attack and dropping the network traffic.

Malicious applications may attempt to detect their operating environment prior to fully executing their payloads. These checks are often used to ensure the application is not running within an analysis environment such as a sandbox used for application vetting, security research, or reverse engineering. Adversaries may use many different checks such as physical sensors, location, and system properties to fingerprint emulators and sandbox environments.[1][2][3][4] Adversaries may access `android.os.SystemProperties` via Java reflection to obtain specific system information.[5] Standard values such as phone number, IMEI, IMSI, device IDs, and device drivers may be checked against default signatures of common sandboxes.[6]

Adversaries may establish persistence and/or elevate privileges using system mechanisms that trigger execution based on specific events. Various operating systems have means to monitor and subscribe to events such as logons or other user activity such as running specific applications/binaries. Cloud environments may also support various functions and services that monitor and can be invoked in response to specific cloud events.[1][2][3]

Adversaries may abuse these mechanisms as a means of maintaining persistent access to a victim via repeatedly executing malicious code. After gaining access to a victim system, adversaries may create/modify event triggers to point to malicious content that will be executed whenever the event trigger is invoked.[4][5][6]

Since the execution can be proxied by an account with higher permissions, such as SYSTEM or service accounts, an adversary may be able to abuse these triggered execution mechanisms to escalate their privileges.

Adversaries may establish persistence using system mechanisms that trigger execution based on specific events. Mobile operating systems have means to subscribe to events such as receiving an SMS message, device boot completion, or other device activities.

Adversaries may abuse these mechanisms as a means of maintaining persistent access to a victim via automatically and repeatedly executing malicious code. After gaining access to a victim’s system, adversaries may create or modify event triggers to point to malicious content that will be executed whenever the event trigger is invoked.

Adversaries may host seemingly genuine Wi-Fi access points to deceive users into connecting to malicious networks as a way of supporting follow-on behaviors such as Network Sniffing, Transmitted Data Manipulation, or Input Capture.[1]

By using a Service Set Identifier (SSID) of a legitimate Wi-Fi network, fraudulent Wi-Fi access points may trick devices or users into connecting to malicious Wi-Fi networks.[2][3] Adversaries may provide a stronger signal strength or block access to Wi-Fi access points to coerce or entice victim devices into connecting to malicious networks.[4] A Wi-Fi Pineapple – a network security auditing and penetration testing tool – may be deployed in Evil Twin attacks for ease of use and broader range. Custom certificates may be used in an attempt to intercept HTTPS traffic.

Similarly, adversaries may also listen for client devices sending probe requests for known or previously connected networks (Preferred Network Lists or PNLs). When a malicious access point receives a probe request, adversaries can respond with the same SSID to imitate the trusted, known network.[4] Victim devices are led to believe the responding access point is from their PNL and initiate a connection to the fraudulent network.

Upon logging into the malicious Wi-Fi access point, a user may be directed to a fake login page or captive portal webpage to capture the victim’s credentials. Once a user is logged into the fraudulent Wi-Fi network, the adversary may able to monitor network activity, manipulate data, or steal additional credentials. Locations with high concentrations of public Wi-Fi access, such as airports, coffee shops, or libraries, may be targets for adversaries to set up illegitimate Wi-Fi access points.

Adversaries who successfully compromise a system may attempt to maintain persistence by “closing the door” behind them – in other words, by preventing other threat actors from initially accessing or maintaining a foothold on the same system.

For example, adversaries may patch a vulnerable, compromised system[1][2] to prevent other threat actors from leveraging that vulnerability in the future. They may “close the door” in other ways, such as disabling vulnerable services[3], stripping privileges from accounts[4], or removing other malware already on the compromised device.[5]

Hindering other threat actors may allow an adversary to maintain sole access to a compromised system or network. This prevents the threat actor from needing to compete with or even being removed themselves by other threat actors. It also reduces the “noise” in the environment, lowering the possibility of being caught and evicted by defenders. Finally, in the case of Resource Hijacking, leveraging a compromised device’s full power allows the threat actor to maximize profit.[3]

Adversaries may execute their own malicious payloads by hijacking the binaries used by an installer. These processes may automatically execute specific binaries as part of their functionality or to perform other actions. If the permissions on the file system directory containing a target binary, or permissions on the binary itself, are improperly set, then the target binary may be overwritten with another binary using user-level permissions and executed by the original process. If the original process and thread are running under a higher permissions level, then the replaced binary will also execute under higher-level permissions, which could include SYSTEM.

Another variation of this technique can be performed by taking advantage of a weakness that is common in executable, self-extracting installers. During the installation process, it is common for installers to use a subdirectory within the %TEMP% directory to unpack binaries such as DLLs, EXEs, or other payloads. When installers create subdirectories and files they often do not set appropriate permissions to restrict write access, which allows for execution of untrusted code placed in the subdirectories or overwriting of binaries used in the installation process. This behavior is related to and may take advantage of DLL search order hijacking.

Adversaries may use this technique to replace legitimate binaries with malicious ones as a means of executing code at a higher permissions level. Some installers may also require elevated privileges that will result in privilege escalation when executing adversary controlled code. This behavior is related to Bypass User Account Control. Several examples of this weakness in existing common installers have been reported to software vendors.[1] [2] If the executing process is set to run at a specific time or during a certain event (e.g., system bootup) then this technique can also be used for persistence.

Adversaries may use execution guardrails to constrain execution or actions based on adversary supplied and environment specific conditions that are expected to be present on the target. Guardrails ensure that a payload only executes against an intended target and reduces collateral damage from an adversary’s campaign.[1] Values an adversary can provide about a target system or environment to use as guardrails may include specific network share names, attached physical devices, files, joined Active Directory (AD) domains, and local/external IP addresses.[2]

Guardrails can be used to prevent exposure of capabilities in environments that are not intended to be compromised or operated within. This use of guardrails is distinct from typical Virtualization/Sandbox Evasion. While use of Virtualization/Sandbox Evasion may involve checking for known sandbox values and continuing with execution only if there is no match, the use of guardrails will involve checking for an expected target-specific value and only continuing with execution if there is such a match.

Adversaries may identify and block certain user-agents to evade defenses and narrow the scope of their attack to victims and platforms on which it will be most effective. A user-agent self-identifies data such as a user's software application, operating system, vendor, and version. Adversaries may check user-agents for operating system identification and then only serve malware for the exploitable software while ignoring all other operating systems.[3]

Adversaries may use execution guardrails to constrain execution or actions based on adversary supplied and environment specific conditions that are expected to be present on the target. Guardrails ensure that a payload only executes against an intended target and reduces collateral damage from an adversary’s campaign. Values an adversary can provide about a target system or environment to use as guardrails may include environment information such as location.[1]

Guardrails can be used to prevent exposure of capabilities in environments that are not intended to be compromised or operated within. This use of guardrails is distinct from typical System Checks. While use of System Checks may involve checking for known sandbox values and continuing with execution only if there is no match, the use of guardrails will involve checking for an expected target-specific value and only continuing with execution if there is such a match.

Adversaries may attempt to leverage Application Program Interfaces (APIs) used for communication between control software and the hardware. Specific functionality is often coded into APIs which can be called by software to engage specific functions on a device or other software.

Adversaries may steal data by exfiltrating it over a different protocol than that of the existing command and control channel. The data may also be sent to an alternate network location from the main command and control server.

Alternate protocols include FTP, SMTP, HTTP/S, DNS, SMB, or any other network protocol not being used as the main command and control channel. Adversaries may also opt to encrypt and/or obfuscate these alternate channels.

Exfiltration Over Alternative Protocol can be done using various common operating system utilities such as Net/SMB or FTP.[1] On macOS and Linux curl may be used to invoke protocols such as HTTP/S or FTP/S to exfiltrate data from a system.[2]

Many IaaS and SaaS platforms (such as Microsoft Exchange, Microsoft SharePoint, GitHub, and AWS S3) support the direct download of files, emails, source code, and other sensitive information via the web console or Cloud API.

Adversaries may steal data by exfiltrating it over a different protocol than that of the existing command and control channel. The data may also be sent to an alternate network location from the main command and control server.

Alternate protocols include FTP, SMTP, HTTP/S, DNS, SMB, or any other network protocol not being used as the main command and control channel. Different protocol channels could also include Web services such as cloud storage. Adversaries may opt to also encrypt and/or obfuscate these alternate channels.

Adversaries may steal data by exfiltrating it over an asymmetrically encrypted network protocol other than that of the existing command and control channel. The data may also be sent to an alternate network location from the main command and control server.

Asymmetric encryption algorithms are those that use different keys on each end of the channel. Also known as public-key cryptography, this requires pairs of cryptographic keys that can encrypt/decrypt data from the corresponding key. Each end of the communication channels requires a private key (only in the procession of that entity) and the public key of the other entity. The public keys of each entity are exchanged before encrypted communications begin.

Network protocols that use asymmetric encryption (such as HTTPS/TLS/SSL) often utilize symmetric encryption once keys are exchanged. Adversaries may opt to use these encrypted mechanisms that are baked into a protocol.

Adversaries may attempt to exfiltrate data over Bluetooth rather than the command and control channel. If the command and control network is a wired Internet connection, an adversary may opt to exfiltrate data using a Bluetooth communication channel.

Adversaries may choose to do this if they have sufficient access and proximity. Bluetooth connections might not be secured or defended as well as the primary Internet-connected channel because it is not routed through the same enterprise network.

Adversaries may steal data by exfiltrating it over an existing command and control channel. Stolen data is encoded into the normal communications channel using the same protocol as command and control communications.

Adversaries may steal data by exfiltrating it over an existing command and control channel. Stolen data is encoded into the normal communications channel using the same protocol as command and control communications.

Adversaries may attempt to exfiltrate data over a different network medium than the command and control channel. If the command and control network is a wired Internet connection, the exfiltration may occur, for example, over a WiFi connection, modem, cellular data connection, Bluetooth, or another radio frequency (RF) channel.

Adversaries may choose to do this if they have sufficient access or proximity, and the connection might not be secured or defended as well as the primary Internet-connected channel because it is not routed through the same enterprise network.

Adversaries may attempt to exfiltrate data over a different network medium than the command and control channel. If the command and control network is a standard Internet connection, the exfiltration may occur, for example, via Bluetooth, or another radio frequency (RF) channel.

Adversaries may choose to do this if they have sufficient access or proximity, and the connection might not be secured or defended as well as the primary Internet-connected channel because it is not routed through the same enterprise network.

Adversaries may attempt to exfiltrate data via a physical medium, such as a removable drive. In certain circumstances, such as an air-gapped network compromise, exfiltration could occur via a physical medium or device introduced by a user. Such media could be an external hard drive, USB drive, cellular phone, MP3 player, or other removable storage and processing device. The physical medium or device could be used as the final exfiltration point or to hop between otherwise disconnected systems.

Adversaries may steal data by exfiltrating it over a symmetrically encrypted network protocol other than that of the existing command and control channel. The data may also be sent to an alternate network location from the main command and control server.

Symmetric encryption algorithms are those that use shared or the same keys/secrets on each end of the channel. This requires an exchange or pre-arranged agreement/possession of the value used to encrypt and decrypt data.

Network protocols that use asymmetric encryption often utilize symmetric encryption once keys are exchanged, but adversaries may opt to manually share keys and implement symmetric cryptographic algorithms (ex: RC4, AES) vice using mechanisms that are baked into a protocol. This may result in multiple layers of encryption (in protocols that are natively encrypted such as HTTPS) or encryption in protocols that not typically encrypted (such as HTTP or FTP).

Adversaries may steal data by exfiltrating it over an un-encrypted network protocol other than that of the existing command and control channel. The data may also be sent to an alternate network location from the main command and control server.[1]

Adversaries may opt to obfuscate this data, without the use of encryption, within network protocols that are natively unencrypted (such as HTTP, FTP, or DNS). This may include custom or publicly available encoding/compression algorithms (such as base64) as well as embedding data within protocol headers and fields.

Adversaries may steal data by exfiltrating it over an un-encrypted network protocol other than that of the existing command and control channel. The data may also be sent to an alternate network location from the main command and control server.

Adversaries may opt to obfuscate this data, without the use of encryption, within network protocols that are natively unencrypted (such as HTTP, FTP, or DNS). Adversaries may employ custom or publicly available encoding/compression algorithms (such as base64) or embed data within protocol headers and fields.

Adversaries may use an existing, legitimate external Web service to exfiltrate data rather than their primary command and control channel. Popular Web services acting as an exfiltration mechanism may give a significant amount of cover due to the likelihood that hosts within a network are already communicating with them prior to compromise. Firewall rules may also already exist to permit traffic to these services.

Web service providers also commonly use SSL/TLS encryption, giving adversaries an added level of protection.