Infrastructure protection - Connected Mobility Lens

Infrastructure protection

CMSEC_13: How do you protect your systems and APIs from unauthorized access or exposure?

[CMSEC_BP13.1] Define a baseline of normal behavior for your vehicle as a reference and reject and alert on any request patterns that deviate.

Implement a solution to create a baseline of metrics for vehicles within your fleet. For example, AWS IoT Device Defender can be used for devices which connect to AWS IoT Core. This baseline can help in detecting anomalies such as impossible motion speeds, or a vehicle existing in multiple locations simultaneously. Through the use of Security Profiles within IoT Core and the enablement of metrics by Device Defender a threshold can be established to define this anomalous behavior. These thresholds can be either rule-based or use Machine Learning to evaluate how these devices should behave. An alert is generated when the defined threshold is crossed which can be used to perform automated mitigation actions you define such as updating a policy or certificate and can also send notifications via Amazon Simple Notification Service (SNS).

CMSEC_14: How do you test Electronic Control Units (ECUs) and implement configuration changes to minimize vulnerabilities?

[CMSEC_BP14.1] Conduct threat modeling using a formal methodology such as the MITRE Threat Assessment and Threat Analysis Risk Assessment (TARA) on in-vehicle systems.

[CMSEC_BP14.2] Conduct tests on ECUs to determine if there are any unnecessary configurations such as open ports, default permissions, unnecessary services running, or other vulnerabilities.

[CMSEC_BP14.3] Develop requirements and processes to address and remediate vulnerabilities discovered through testing.

CMSEC_15: How do you manage processes for updating the software of components within your connected vehicles?

CMSEC_BP15.1] Validate all software updates are cryptographically signed upon release and verified before installation.

For devices which use AWS IoT, software updates can be signed using AWS Signer a fully managed code-signing service. AWS Signer uses certificates imported to AWS Certificate Manager to sign software, AWS Identity Access Management (IAM) to assign roles to restrict who can sign and in what Region. All actions within AWS Signer are also captured in AWS CloudTrail for auditing purposes.

[CMSEC_BP15.2] Test software packages for vulnerabilities and errors prior to deployment.

[CMSEC_BP15.3] Perform controlled and secure deployment of software to vehicle fleets with rollback capabilities in case of failures.

CMSEC_16: How do you maintain an accurate inventory of your connected vehicles and associated software or hardware components?

[CMSEC_BP16.1] Maintain an accurate and continuously updated inventory of vehicles and any versions of software packages or underlying hardware.

This should include any applicable Software Bill of Material (SBOM) documents as outlined by something such as The Minimum Elements For a Software Bill of Materials (SBOM) pursuant to Executive Order 14028 including any Vulnerability Exploitability eXchange (VEX) documents as outlined by the United States Department of Energy in coordination with the National Telecommunications and Information Administration. A VEX document is an attestation and contains information about a product and whether it is affected by a known vulnerability.

CMSEC_17: How do you verify the integrity of the software running within your vehicles?

[CMSEC_BP17.1] Use methods for ensuring integrity such as through secure boot and software signing methodologies.

CMSEC_18: How do you secure the Software Development Lifecycle (SDLC)?

[CMSEC_BP18.1] Develop SDLC policies which govern these requirements and provide prescriptive guidance for implementing required controls.

CMSEC_19: How do you manage vulnerabilities for your connected vehicles?

[CMSEC_BP19.1] Continuously monitor informational feeds on threat intelligence from vendors, suppliers, third-party intelligence providers, repositories, and communities.

Threat intelligence feeds provide additional insights into signatures of potentially malicious activity such as malware, botnets, and phishing. This information helps to supplement the tools already in use by security teams to understand the current threat landscape at a broader scope. AWS Partners such as Recorded Future and Tenable provide threat intelligence feeds and are readily available through the AWS Marketplace. IP based threat lists from these providers can also be used to supplement detections within AWS GuardDuty our purpose-built threat detection service. Customers can also leverage an AWS partner like Argus Cyber Security to help implement in-vehicle security systems like an Intrusion Detection System for Controller Area Networks (CAN), vehicle vulnerability management, and firewall capabilities to protect components. Industry specific feeds such as Auto-ISAC can provide additional insights based on contributions from organizations within automotive.

[CMSEC_BP19.2] Conduct vulnerability impact analysis and testing in isolated and secure environments.

When conducting tests of vulnerabilities on systems or investigating potentially compromised systems an isolated and secure environment should be used. This verifies that other resources are not inadvertently impacted or compromised by these activities. When conducting this testing in AWS environments a separate Account should be created to provide segmentation from other Accounts and resources. Strong defense-in-depth configurations should also be used such as for forensic investigations. Some of these strategies are outlined in the Forensic investigation environment strategies in the AWS Cloud blog post.

[CMSEC_BP19.3] Define a process for classifying and prioritizing vulnerabilities for remediation.

Processes for classification, prioritization, and remediation of vulnerabilities should be formally documented and readily accessible to stakeholders. Playbooks and runbooks specific to your organizations technology stack and operational processes should be used as a means of automating segments of the investigation and help minimize manual errors during a potentially stressful situation. Automation can be accomplished through solutions such as Jupyter Notebooks or running automation documents within AWS Systems Manager for example.