Study guide

1.2 Amazon GuardDuty Architecture & Finding Types

AWS Certified Security – Specialty SCS-C03

Track your progress as you read

Finished sections get a check mark.

1 Chapter 0: Getting Started with the SCS-C03 Exam

0.1 SCS-C03 Exam Format, Domains & Study Strategy

2 Chapter 1: Security Monitoring, Alerting & Threat Detection Services

1.1 Workload Monitoring & Health Check Strategies 1.2 Amazon GuardDuty Architecture & Finding Types 1.3 AWS Security Hub & Automated Conformance Assessment 1.4 Amazon Macie & Automated Data Classification

3 Chapter 2: Enterprise Logging Architecture & Analysis

2.1 Multi-Account CloudTrail & Organization Trails 2.2 Amazon Security Lake & OCSF Log Normalization 2.3 Network Telemetry: VPC Flow Logs & Route 53 Resolver Logs 2.4 Troubleshooting Security Telemetry & Log Ingestion 2.5 Analyzing, Normalizing & Correlating Security Logs

4 Chapter 3: Incident Readiness, Response Planning & Automated Runbooks

3.1 Incident Response Planning & Blast Radius Mitigation 3.2 Testing & Validating Incident Response Plans 3.3 Automated Incident Remediation with EventBridge & Step Functions 3.4 Systems Manager OpsCenter & Incident Manager Runbooks

5 Chapter 4: Digital Forensics, Evidence Preservation & Root Cause Analysis

4.1 Forensic Artifact Capture & Chain of Custody 4.2 Threat Containment & Network Quarantine Techniques 4.3 Correlating Events & Root Cause Analysis with Amazon Detective

6 Chapter 5: Edge Protection, WAF & DDoS Mitigation

5.1 CloudFront Security Headers & Origin Access Control 5.2 AWS WAF Architecture & Advanced Rule Sets 5.3 AWS Shield Advanced & DDoS Defense Architectures 5.4 Edge Integrations: Managed Rule Groups, OCSF Ingestion, IoT Policies & CORS

7 Chapter 6: Compute Workload Hardening, Vulnerability Management & GenAI Security

6.1 Golden AMIs & Container Image Hardening Pipelines 6.2 Amazon Inspector Vulnerability Management & Patch Automation 6.3 Generative AI Security: Amazon Bedrock Guardrails & Model Protection 6.4 Instance Profiles, Service Roles & Execution Roles for Compute 6.5 Secure Administrative Access: Session Manager & EC2 Instance Connect

8 Chapter 7: Network Security, Segmentation, Private Connectivity & Encryption in Transit

7.1 Layered Traffic Filtering: SGs, NACLs & AWS Network Firewall 7.2 Hybrid Connectivity Security: VPN, Direct Connect & MACsec 7.3 Zero-Trust Access & Network Access Analysis 7.4 Enforcing Encryption in Transit: ELB Security Policies, TLS & HTTPS-Only Access 7.5 Private Access: VPC Endpoints, PrivateLink, Client VPN & Verified Access 7.6 Inter-Resource Encryption in Transit: Nitro, EMR, EKS & SageMaker AI

9 Chapter 8: Enterprise Authentication, Federation & Directory Services

8.1 AWS IAM Identity Center & Enterprise IdP Integration 8.2 Amazon Cognito User Pools & Identity Pools 8.3 AWS Security Token Service (STS) & Temporary Credential Workflows 8.4 Troubleshooting Identity & Authentication Failures

10 Chapter 9: Advanced Authorization Architecture & Policy Evaluation

9.1 Complete AWS Policy Evaluation Logic & Decision Flow 9.2 Permission Boundaries & Safe Admin Delegation 9.3 Attribute-Based Access Control (ABAC) with IAM Tags

11 Chapter 10: Least Privilege, Policy Validation & External Access Auditing

10.1 IAM Roles Anywhere & Amazon Verified Permissions 10.2 Auditing External & Unused Access with IAM Access Analyzer 10.3 Analyzing Authorization Failures & Policy Simulation

12 Chapter 11: Cryptographic Services: KMS Architecture & Key Governance

11.1 AWS KMS Architecture, Key Policies & Key Types 11.2 KMS Grants, Encryption Context & Multi-Region Keys 11.3 Imported Key Material (BYOK) & CloudHSM Key Stores

13 Chapter 12: Secrets Management, Sensitive Data Masking & PKI

12.1 AWS Secrets Manager & Systems Manager Parameter Store 12.2 Sensitive Data Masking in Telemetry & Messaging 12.3 AWS Certificate Manager (ACM) & Private CA Architecture

14 Chapter 13: Storage Encryption, Integrity, Lifecycle & Backup Protection

13.1 S3 Encryption Modes & S3 Bucket Keys 13.2 Data Integrity Protection: Object Lock, Vault Lock & Code Signing 13.3 Secure Data Lifecycle, Cross-Account Replication & Backup Vault Lock 13.4 Lifecycle, Backup & Secure Replication Beyond S3: EFS, FSx, DLM, DataSync & Air-Gapped Vaults

15 Chapter 14: Multi-Account Architecture, Organizations & Control Tower Governance

14.1 AWS Organizations Structure, Hierarchy & Resource Control 14.2 AWS Control Tower Landing Zone & Guardrail Framework 14.3 Delegated Administrator Architecture & Root User Governance 14.4 Centralized Resource Sharing with RAM & Service Catalog

16 Chapter 15: Compliance Automation, Infrastructure as Code Security & Auditing

15.1 Infrastructure as Code Security & Multi-Account Deployment 15.2 AWS Config, Conformance Packs & Automated Remediation 15.3 Audit Evidence Collection & Well-Architected Security Reviews

1.2 Amazon GuardDuty Architecture & Finding Types

Key Takeaways Explain This Topic AI

GuardDuty is an agentless, continuous threat detection service that analyzes AWS control plane and network telemetry out-of-band without customer-incurred VPC Flow Log ingestion fees or workload latency.

GuardDuty's three foundational data sources are AWS CloudTrail management events, VPC Flow Logs, and Route 53 Resolver DNS query logs; S3 data events and EKS audit logs are analyzed only when the S3 Protection and EKS Protection plans are enabled.

Specialized protection plans extend detection to managed workloads: S3 Protection, EKS Protection, RDS Protection, Lambda Protection, and Runtime Monitoring via an in-guest eBPF agent.

Findings follow the [ThreatFamily]:[TargetResource]/[ThreatSubfamily].[Artifact] taxonomy and carry a severity from 1.0 to 10.0: Low (1.0–3.9), Medium (4.0–6.9), High (7.0–8.9), and Critical (9.0–10.0) for attack-sequence findings.

Suppression rules auto-archive expected findings; suppressed findings stay queryable in GuardDuty but are not sent to Security Hub CSPM, the S3 findings export, Detective, or EventBridge.

Last updated: September 2026

1.2 Amazon GuardDuty Architecture & Finding Types

Amazon GuardDuty is a managed threat detection service that continuously monitors AWS accounts, workloads, and data for malicious activity and unauthorized behavior. Rather than relying solely on static rule sets, GuardDuty combines machine learning, anomaly detection, behavioral modeling, and integrated threat intelligence (including AWS Security Intelligence and third-party feeds such as Proofpoint and CrowdStrike) to identify indicators of compromise (IoCs).

Understanding GuardDuty's architectural foundations, out-of-band telemetry processing, specialized protection plans, finding format, and multi-account administration is essential for the AWS Certified Security – Specialty exam.

GuardDuty Core Architecture & Out-of-Band Telemetry Ingestion

A defining architectural characteristic of GuardDuty is its out-of-band, serverless operation . Enabling GuardDuty does not deploy software agents on EC2 instances (for foundational features), does not proxy network traffic, and does not introduce latency or resource overhead to production workloads.

Loading diagram...

Foundational Data Sources

When GuardDuty is enabled, it automatically begins consuming three foundational data sources (CloudTrail management events, VPC Flow Logs, and Route 53 Resolver DNS query logs) without requiring the customer to create or configure them. Items 2 and 5 below are analyzed only when their optional protection plans are turned on:

AWS CloudTrail Management Events : Tracks API calls across all AWS services, identifying reconnaissance (e.g., automated Describe* sweeps), unauthorized IAM policy changes, and credential access.

CloudTrail S3 Data Events (S3 Protection plan): Analyzes object-level operations ( GetObject , PutObject , DeleteObject ) to detect anomalous data access patterns and mass downloads.

VPC Flow Logs : Analyzes IP traffic metadata to and from EC2 network interfaces. Critical Exam Fact : GuardDuty consumes VPC Flow Log telemetry through an independent, duplicated stream that AWS manages. The customer does not need to enable VPC Flow Logs in their VPC to use GuardDuty, nor do they pay CloudWatch Logs ingestion or storage fees for GuardDuty's internal consumption.

Route 53 DNS Query Logs : Analyzes DNS queries sent to the default AWS Route 53 Resolver (the VPC +2 address). Detects compromised instances querying known Command-and-Control (C2) domains, dynamic DNS services, or domain generation algorithms (DGA).

Kubernetes (EKS) Audit Logs (EKS Protection plan): Consumes audit logs directly from the managed Amazon EKS control plane to detect cluster role binding escalations, anonymous API requests, and unauthorized pod launches.

Specialized & Advanced Protection Plans

Beyond foundational log analysis, GuardDuty provides specialized protection features that target specific workload vectors. These features can be enabled independently within the GuardDuty console or via the UpdateDetector API.

Protection Feature Telemetry Analyzed Workload Impact & Agent Security Detection Scope

S3 Protection S3 data plane API events Agentless; out-of-band Anomalous access volume, requests from malicious IPs or Tor nodes, exfiltration.

EKS Protection Amazon EKS audit logs Agentless; control plane Kubernetes control plane privilege escalation, malicious pod creation, compromised service accounts.

Runtime Monitoring OS system calls, processes, socket activity eBPF security agent on host/pod Fileless malware, unauthorized process execution, reverse shells, container breakouts.

RDS Protection Aurora login activity Agentless; database control plane Brute-force login attempts, anomalous client IPs, credential stuffing against Aurora MySQL/PostgreSQL.

Lambda Protection VPC flow metadata from Lambda invocations Agentless; network layer Compromised serverless functions communicating with C2 nodes or cryptocurrency mining pools.

Malware Protection EBS volume snapshots / S3 objects Agentless; isolated scan environment Scans EBS volumes attached to infected EC2 instances and newly uploaded S3 objects for trojans, worms, and rootkits.

Deep Dive: GuardDuty Runtime Monitoring & eBPF Architecture

While foundational GuardDuty relies on external network and control plane logs, GuardDuty Runtime Monitoring provides deep in-guest operating system visibility. It supports Amazon EKS , Amazon ECS (including AWS Fargate), and Amazon EC2 .

Runtime Monitoring operates using an extended Berkeley Packet Filter (eBPF) security agent:

Automated Agent Management : When enabled for EKS, GuardDuty can automatically deploy and update the security agent daemonset ( aws-guardduty-agent ) across cluster worker nodes.

Kernel-Level Telemetry : The eBPF agent intercepts operating system system calls ( execve , connect , open ) in kernel space without modifying application source code.

Process Context : Identifies the specific Linux process name, parent process ID, container ID, and command-line arguments responsible for malicious outbound connections or file modifications.

Deep Dive: Malware Protection Mechanics

When GuardDuty generates a compute-based finding (such as an EC2 instance beaconing to a known C2 server), GuardDuty Malware Protection can automatically initiate an agentless scan:

GuardDuty creates a point-in-time Amazon EBS snapshot of the target instance's root and attached volumes.

GuardDuty shares the snapshot with an isolated AWS-managed service account in the same Region.

An ephemeral scanning instance in the service account restores the snapshot to an encrypted EBS volume and performs signature and heuristic malware scanning.

When scanning completes, GuardDuty automatically deletes the restored volume and snapshot (retention can optionally be enabled for forensic examination).

GuardDuty Finding Taxonomy & Severity Scoring

GuardDuty findings represent potential security incidents formatted as structured JSON payloads. Every finding follows a standardized three-part naming hierarchy: [ThreatFamily]:[TargetResource]/[ThreatSubfamily].[Artifact]

Finding Taxonomy Components

ThreatFamily : The broad category of threat or attacker motivation (e.g., CryptoCurrency , UnauthorizedAccess , Recon , Stealth , Trojan , PrivilegeEscalation , Impact ).

TargetResource : The AWS resource targeted or compromised (e.g., EC2 , IAMUser , S3 , Kubernetes , RDS , Lambda , Runtime ).

ThreatSubfamily : The specific attack technique, malware family, or anomalous behavior (e.g., SSHBruteForce , BitcoinTool.B , TorIPCaller , DNSDataExfiltration , UnusualDoHTraffic ).

Artifact : Optional detail describing the data source or communication channel through which the threat was detected (e.g., !DNS , !VPC ).

Example Finding : CryptoCurrency:EC2/BitcoinTool.B!DNS denotes that an EC2 instance was detected querying known cryptocurrency mining pool endpoints via DNS.

Severity Scoring System

GuardDuty assigns every finding a numeric severity value between 1.0 and 10.0 . Findings fall into four operational bands:

Severity Band Numeric Score Range Operational Meaning & Typical Scenarios Recommended Response SLA

Low 1.0 - 3.9 Suspicious activity detected, but no evidence of successful compromise (e.g., external port scanning targeting an EC2 instance where security groups blocked the traffic; minimal anomalous API volume). Log for audit; review in weekly triage.

Medium 4.0 - 6.9 Suspicious activity that deviates from historical baseline and warrants prompt investigation (e.g., API calls executed from a Tor exit node; unusual spike in S3 object downloads from a corporate network). Investigate within 4–8 hours.

High 7.0 - 8.9 Active, verified compromise or high-impact threat in progress (e.g., EC2 instance communicating with known C2 botnet; IAM access keys used to create backdoor users; ransomware file encryption patterns). Immediate automated response and SOC paging (< 15 minutes).

Critical 9.0 - 10.0 An attack sequence may be in progress or has recently happened. GuardDuty Extended Threat Detection correlates several signals (for example, credential misuse followed by S3 data exfiltration) into one attack-sequence finding. Treat as a declared incident; page responders immediately.

Exam Tip : The GuardDuty severity scale now runs to 10.0 ; Critical (9.0–10.0) findings come from Extended Threat Detection attack sequences. An EventBridge condition of "severity": [{"numeric": [">=", 7]}] therefore captures both High and Critical findings.

Suppression Rules & Finding Lifecycle

In enterprise environments, authorized activities can trigger false-positive findings. Examples include authorized penetration testing, vulnerability scanning appliances, internal CI/CD pipelines deploying ephemeral nodes, and third-party SaaS management tools.

How Suppression Rules Function

Rather than disabling detection rules globally, security engineers configure Suppression Rules :

Suppression rules define precise filter criteria based on finding attributes (such as type , resource.tag , service.action.awsApiCallAction.remoteIpDetails.ipAddressV4 , or accountId ).

When a finding matches an active suppression rule, GuardDuty automatically sets the finding's RecordState to ARCHIVED upon generation.

Audit Preservation : Suppressed findings are not discarded . They are stored in the GuardDuty repository for 90 days , where they can be queried via the console or CLI by filtering for archived findings. However, suppressed findings are not sent to Security Hub CSPM, the S3 findings export, Amazon Detective, or Amazon EventBridge, and they do not start Malware Protection for EC2 scans. A suppression rule is therefore also a routing decision: write it narrowly.

Multi-Account Governance via AWS Organizations

Managing GuardDuty across hundreds of AWS accounts requires centralized governance. AWS Organizations enables a Delegated Administrator model that separates security operations from the AWS Organizations management account.

Loading diagram...

Delegated Administrator Capabilities & Constraints

Designation : The Organizations management account calls EnableOrganizationAdminAccount designating a dedicated security tooling account as the GuardDuty Delegated Administrator.

Auto-Enablement : The delegated administrator enables GuardDuty across all existing member accounts with a single operation and activates Auto-Enable , ensuring that any newly created or joined accounts in the organization immediately have GuardDuty and specialized protection features activated.

Centralized Visibility : The delegated administrator views, searches, and manages findings across all member accounts from a single pane of glass.

Member Account Restrictions : Member accounts cannot disable GuardDuty, cannot remove themselves from the organization detector, and cannot alter suppression rules published by the delegated administrator.

Long-Term Export : GuardDuty findings can be exported to a central Amazon S3 bucket located in the security tooling account. The export destination requires an S3 bucket policy granting guardduty.amazonaws.com permission to put objects and a KMS Key Policy granting GuardDuty permission to use a customer-managed key (CMK) for envelope encryption.

Specialty Exam Pitfalls & Threat Detection Scenarios

The VPC Flow Log Requirement Myth : A frequent exam trap suggests that a customer must create a CloudWatch VPC Flow Log subscription before GuardDuty can analyze network traffic. This is false: GuardDuty taps internal AWS network telemetry independently. Enabling or disabling customer VPC Flow Logs has zero impact on GuardDuty's operational coverage.

KMS Permissions for Finding Export : When configuring GuardDuty finding export to an S3 bucket, using the AWS-managed key aws/s3 will fail. GuardDuty requires a customer-managed KMS key (CMK) because the GuardDuty service principal must be granted explicit kms:GenerateDataKey permissions via the KMS key policy.

Regional Isolation : GuardDuty is a Regional service. Enabling GuardDuty in us-east-1 provides zero threat detection for workloads operating in eu-west-1 or ap-southeast-1 . To protect an enterprise, GuardDuty must be enabled in all active and inactive AWS Regions (to detect unauthorized resource creation in unused Regions).

Suppression Rules vs EventBridge Filters : Suppression rules archive findings within the GuardDuty service itself, preventing them from cluttering the console or feeding default Security Hub compliance checks. If you only filter events in EventBridge, the findings still remain active in the GuardDuty console and generate alerts in integrated downstream services.

Loading diagram... Amazon GuardDuty Threat Ingestion, Processing & Automated Response