Tags: Cyber Security »»»» Energy Infrastructure
The Cybersecurity and Infrastructure Security Agency announced today it had recently worked on a cybersecurity threat to the operational network of a natural gas compression facility. Such facilities compress natural gas, and presumably the CISA is referring to a facility that prepares CNG for sale to other countries. The cyber-attack caused a couple days of outage, and the facility lost revenue for those two days. While there was no immediate threat -- the attacker had access to the control system, and instead of leaving behind a ransomware attack could have caused damage to systems.
The press release (see below) says the attacker launched the attack using a Spearphishing Link. What that means is the attacker sent an email to an employee of the site, and the email contained a link that took the recipient to a bogus site. And by getting someone to visit that bogus site, the Spearfishing Attacker was able to gain access to the internal networks.
In this case the Spearfishing attack gave the attacker access to the Information Technology network -- meaning the office network.
Through some undisclosed means, the attacker then gained access to the Operational Network of the site. The operational network contains the computers used by engineers to control the machinery on the site.
The attacker then deployed an off-the-shelf ransomware attack - I believe such attacks encrypt the contents of computer systems, and leave behind a note describing how to pay off the attacker so the machines will be decrypted.
The affected systems are described as human machine interfaces (HMIs), data historians, and polling servers. The impact was that machines weren't able to read the operational data, and store it for analysis (aggregation). The programmable logic controllers (PLCs) were not affected.
Those buzzwords mean:
- human machine interfaces -- This refers to the information displays, knobs, dials, and other control thingies, that the humans interact with in order to control the system.
- data historians -- As the name implies, this keeps track of data that comes from sensors.
- polling servers -- Likewise, this equipment I believe also collects data.
- programmable logic controllers -- These are embedded control devices that are frequently used in industrial control applications. PLC's are programmed using logic that's similar to software, and the kinds of PLC include sensors to read data, or various kinds of actuators to control machines.
In such an environment - PLC's will be scattered among the machines that make up the plant. The PLC's are wired back to the control room and use data communications to send data and commands back and forth. When an engineer in the control room turns a knob, it causes commands to be sent over the wires to the associated PLC device to cause some action like open or close a valve.
Source:
https://www.us-cert.gov/ncas/alerts/aa20-049a
AA20-049A: Ransomware Impacting Pipeline Operations
02/18/2020 08:06 AM EST
Original release date: February 18, 2020
Summary Note: This Activity Alert uses the MITRE Adversarial Tactics, Techniques, and Common Knowledge (ATT&CK™) framework. See the MITRE ATT&CK for Enterprise and ATT&CK for Industrial Control Systems (ICS) frameworks for all referenced threat actor techniques and mitigations.
This Activity Alert summarizes an incident to which CISA recently responded. It is being shared publicly to promote awareness and encourage mitigations by asset owner operators across all critical infrastructure sectors.
The Cybersecurity and Infrastructure Security Agency (CISA) responded to a cyberattack affecting control and communication assets on the operational technology (OT) network of a natural gas compression facility. A cyber threat actor used a Spearphishing Link [T1192] to obtain initial access to the organization’s information technology (IT) network before pivoting to its OT network. The threat actor then deployed commodity ransomware to Encrypt Data for Impact [T1486] on both networks. Specific assets experiencing a Loss of Availability [T826] on the OT network included human machine interfaces (HMIs), data historians, and polling servers. Impacted assets were no longer able to read and aggregate real-time operational data reported from low-level OT devices, resulting in a partial Loss of View [T829] for human operators. The attack did not impact any programmable logic controllers (PLCs) and at no point did the victim lose control of operations. Although the victim’s emergency response plan did not specifically consider cyberattacks, the decision was made to implement a deliberate and controlled shutdown to operations. This lasted approximately two days, resulting in a Loss of Productivity and Revenue [T828], after which normal operations resumed. CISA is providing this Alert to help administrators and network defenders protect their organizations against this and similar ransomware attacks.
Technical Details
Network and Assets
- The victim failed to implement robust segmentation between the IT and OT networks, which allowed the adversary to traverse the IT-OT boundary and disable assets on both networks.
- The threat actor used commodity ransomware to compromise Windows-based assets on both the IT and OT networks. Assets impacted on the organization’s OT network included HMIs, data historians, and polling servers.
- Because the attack was limited to Windows-based systems, PLCs responsible for directly reading and manipulating physical processes at the facility were not impacted.
- The victim was able to obtain replacement equipment and load last-known-good configurations to facilitate the recovery process.
- All OT assets directly impacted by the attack were limited to a single geographic facility.
Planning and Operations
- At no time did the threat actor obtain the ability to control or manipulate operations. The victim took HMIs that read and control operations at the facility offline. A separate and geographically distinct central control office was able to maintain visibility but was not instrumented for control of operations.
- The victim’s existing emergency response plan focused on threats to physical safety and not cyber incidents. Although the plan called for a full emergency declaration and immediate shutdown, the victim judged the operational impact of the incident as less severe than those anticipated by the plan and decided to implement limited emergency response measures. These included a four-hour transition from operational to shutdown mode combined with increased physical security.
- Although the direct operational impact of the cyberattack was limited to one control facility, geographically distinct compression facilities also had to halt operations because of pipeline transmission dependencies. This resulted in an operational shutdown of the entire pipeline asset lasting approximately two days.
- Although they considered a range of physical emergency scenarios, the victim’s emergency response plan did not specifically consider the risk posed by cyberattacks. Consequently, emergency response exercises also failed to provide employees with decision-making experience in dealing with cyberattacks.
- The victim cited gaps in cybersecurity knowledge and the wide range of possible scenarios as reasons for failing to adequately incorporate cybersecurity into emergency response planning.
Mitigations
Asset owner operators across all sectors are encouraged to consider the following mitigations using a risk-based assessment strategy.
Planning and Operational Mitigations
- Ensure the organization’s emergency response plan considers the full range of potential impacts that cyberattacks pose to operations, including loss or manipulation of view, loss or manipulation of control, and loss of safety. In particular, response playbooks should identify criteria to distinguish between events requiring deliberate operational shutdown versus low-risk events that allow for operations to continue.
- Exercise the ability to fail over to alternate control systems, including manual operation while assuming degraded electronic communications. Capture lessons learned in emergency response playbooks.
- Allow employees to gain decision-making experience via tabletop exercises that incorporate loss of visibility and control scenarios. Capture lessons learned in emergency response playbooks.
- Identify single points of failure (technical and human) for operational visibility. Develop and test emergency response playbooks to ensure there are redundant channels that allow visibility into operations when one channel is compromised.
- Implement redundant communication capabilities between geographically separated facilities responsible for the operation of a single pipeline asset. Coordinate planning activities across all such facilities.
- Recognize the physical risks that cyberattacks pose to safety and integrate cybersecurity into the organization’s safety training program.
- Ensure the organization’s security program and emergency response plan consider third parties with legitimate need for OT network access, including engineers and vendors.
Technical and Architectural Mitigations
- Implement and ensure robust Network Segmentation [M1030] between IT and OT networks to limit the ability of adversaries to pivot to the OT network even if the IT network is compromised. Define a demilitarized zone (DMZ) that eliminates unregulated communication between the IT and OT networks.
- Organize OT assets into logical zones by taking into account criticality, consequence, and operational necessity. Define acceptable communication conduits between the zones and deploy security controls to Filter Network Traffic [M1037] and monitor communications between zones. Prohibit Industrial Control System (ICS) protocols from traversing the IT network.
- Require Multi-Factor Authentication [M1032] to remotely access the OT and IT networks from external sources.
- Implement regular Data Backup [M1053] procedures on both the IT and OT networks. Ensure that backups are regularly tested and isolated from network connections that could enable the spread of ransomware.
- Ensure user and process accounts are limited through Account Use Policies [M1036], User Account Control [M1052], and Privileged Account Management [M1026]. Organize access rights based on the principles of least privilege and separation of duties.
- Enable strong spam filters to prevent phishing emails from reaching end users. Implement a User Training [M1017] program to discourage users from visiting malicious websites or opening malicious attachments. Filter emails containing executable files from reaching end users.
- Filter Network Traffic [M1037] to prohibit ingress and egress communications with known malicious Internet Protocol (IP) addresses. Prevent users from accessing malicious websites using Uniform Resource Locator (URL) blacklists and/or whitelists.
- Update Software [M1051], including operating systems, applications, and firmware on IT network assets. Use a risk-based assessment strategy to determine which OT network assets and zones should participate in the patch management program. Consider using a centralized patch management system.
- Set Anti-virus/Anti-malware [M1049] programs to conduct regular scans of IT network assets using up-to-date signatures. Use a risk-based asset inventory strategy to determine how OT network assets are identified and evaluated for the presence of malware.
- Implement Execution Prevention [M1038] by disabling macro scripts from Microsoft Office files transmitted via email. Consider using Office Viewer software to open Microsoft Office files transmitted via email instead of full Microsoft Office suite applications.
- Implement Execution Prevention [M1038] via application whitelisting, which only allows systems to execute programs known and permitted by security policy. Implement software restriction policies (SRPs) or other controls to prevent programs from executing from common ransomware locations, such as temporary folders supporting popular internet browsers or compression/decompression programs, including the AppData/LocalAppData folder.
- Limit Access to Resources over Network [M1035], especially by restricting Remote Desktop Protocol (RDP). If after assessing risks RDP is deemed operationally necessary, restrict the originating sources and require Multi-Factor Authentication [M1032].
