Selective CloudTrail Data Events Logging with Cost Guardrails
Specification & Description
Enable CloudTrail Data event logging for sensitive data stores (S3 objects, DynamoDB tables, and Lambda invocations) using Advanced Event Selectors while filtering out benign AWS internal service events to control ingestion costs.
Technical Specifications
While AWS CloudTrail Management Events capture control-plane operations (e.g., CreateBucket, AttachRolePolicy),
they do not record data-plane operations (e.g., S3 GetObject, PutObject, DeleteObject, DynamoDB GetItem, or Lambda Invoke).
1. High-Value Data Plane Auditing
- Data events provide granular forensic trails required to detect unauthorized data exfiltration, insider threats, and ransomware encryption attempts.
- Organizations must enable Data event logging on S3 buckets containing sensitive data (PII, financial data, secret stores) and production DynamoDB tables.
2. Cost & Volume Management via Advanced Event Selectors
- Unfiltered Data event logging across high-throughput data lakes or transient utility buckets causes exponential CloudTrail ingestion and S3 storage costs ($0.10 per 100,000 data events).
- CloudTrail Advanced Event Selectors must be utilized to:
- Target specific sensitive resource ARNs or prefix patterns.
- Explicitly exclude AWS service-initiated events (
readOnly = trueoreventType = AwsServiceEvent) to eliminate repetitive internal read noise. - Separate read-only audits from write-only modifications where appropriate.
Security Rationale
Management events show who modified infrastructure, but Data events prove what data was actually accessed, read, or exfiltrated. Without data-plane audit trails on sensitive buckets and databases, security teams cannot determine the blast radius or data loss impact of compromised credentials. Filtering service noise ensures comprehensive security visibility without uncontrolled cost surges.
Business & Security Impact
Absence of data event logging creates blind spots in data access patterns, leaving unauthorized downloads or deletions undetectable. Conversely, unconstrained logging across all buckets causes massive log fatigue and severe unexpected billing spikes.
Related Controls & Guardrails (3)
Compliance Framework Mappings (5)
Regulatory baselines and industry standards mapped to this control specification.
| Framework / Standard | Mapped Requirements & Controls | Controls |
|---|---|---|
NIST 800-53 Rev 5NIST 800-53 | AU-2AU-3AU-12AC-3 | 4 |
ISO 27001:2022ISO 27001 | A.8.15A.8.12 | 2 |
SOC 2SOC 2 | CC6.1CC7.2 | 2 |
PCI DSS v4.0PCI DSS | 10.2.110.2.2 | 2 |
HIPAA Security RuleHIPAA | 164.312(b) | 1 |
* Compliance framework mappings are provided for informational architectural reference. Always verify control alignment against your organization's formal audit requirements.
Automated Detection & Rule Logic
Detection Engine: aws-api • Automated: Yes
Inspects organization trails to verify that Advanced Event Selectors or Event Selectors are configured to capture Data events for sensitive resource types (AWS::S3::Object, AWS::DynamoDB::Table).
Preventative Enforcement Policies
Remediation Guide & Automation
Configure CloudTrail Advanced Event Selectors from the AWS Organizations Management account or delegated administrator to record data plane events for sensitive S3 buckets, DynamoDB tables, and Lambda functions while filtering out routine AWS internal service events.
Step-by-Step Instructions
- Identify all sensitive S3 buckets, DynamoDB tables, and critical Lambda functions across member accounts.
- Create an Advanced Event Selector configuration JSON targeting those resource ARNs.
- Apply the selectors to the organization trail targeting the centralized Log Archive bucket.
- Add exclusions for benign internal AWS service events (e.g. eventType != AwsServiceEvent).
- Verify data event delivery in the central Log Archive S3 bucket.