Easy ReviewDOP-C02
Configuration Management and IaC34

GuardDuty

Detect threats in AWS activity and route findings for investigation and response.

Este conteúdo de estudo ainda não foi traduzido, então aparece em inglês. Ajude a traduzir

GuardDuty analyzes account and workload signals for potential threats. Its findings can drive notifications and response workflows across an organization.

GuardDuty basics

Amazon GuardDuty watches account activity and logs for signs of attack and reports findingsA finding is a report of a potential security issue detected in an AWS account, workload, or data..

How it works

  1. Turn on GuardDuty in the account.
  2. It analyzes AWS logs and data sources continuously.
  3. It reports findings such as compromised credentials or unusual access.

Exam cues

  • If the question says intelligent threat detection or compromised credentials, think GuardDuty.

Remember

  • Continuous threat detection with threat intelligence and machine learning.

Trap: GuardDuty detects threats; Inspector finds vulnerabilities; Macie finds sensitive data in S3.

Da documentação da AWS (4)
  • Amazon GuardDuty is a threat detection service that continuously analyzes AWS data sources and logs, using threat intelligence and machine learning, to find malicious activity such as compromised credentials.
  • GuardDuty analyzes AWS data sources using threat intelligence and machine learning to identify suspicious and potentially malicious activity.
  • GuardDuty can detect compromised credentials and unauthorized cryptomining on EC2 instances and container workloads.
  • GuardDuty automatically ingests CloudTrail management events, EC2 VPC flow logs, and DNS logs as foundational data sources.

Data sources and threat findings

Amazon GuardDuty detects potential threats by analyzing AWS activity and threat intelligence. A findingA finding is a report of a potential security issue detected in an AWS account, workload, or data. is a report of a potential security issue, not proof that a resource is compromised.

How it works

  1. Enable GuardDuty in a Region. It starts analyzing CloudTrail management eventsA management event records a management operation performed on an AWS resource., EC2 VPC flow logsA VPC flow log captures information about IP traffic to and from network interfaces., and DNS logs.
  2. GuardDuty receives independent streams. It does not need a customer-configured trail or VPC flow log. It extracts fields for analysis and discards the source logs, so configure your own logging when you need retention.
  3. GuardDuty compares activity with threat intelligence, such as known malicious addresses, and uses machine learning to identify anomalies. It creates findings for potential threats, including compromised credentials and unauthorized cryptomining.
  4. Choose protection plans for other signals. An audit log records actions.
  5. Configure the same delegated administratorA delegated administrator is an account designated to manage GuardDuty for associated member accounts. and members in each desired Region. GuardDuty processes relevant global service events in each enabled Region.
  6. Route active findings through EventBridge or configure S3 export for longer retention. S3 export requires bucket and KMS key policies; restrict access using the source account and detector ARN.
Protection What it analyzes
S3 Protection S3 data events for risks such as exfiltration, the unauthorized movement of data
EKS Audit Log Monitoring Kubernetes audit logs
Malware Protection EC2 EBS volumes or newly uploaded S3 objects, depending on the plan

Exam cues

  • If the question says unusual API calls or cryptomining, think GuardDuty, because it identifies potentially malicious activity from AWS signals.
  • If the question says no trail or VPC flow log delivery exists, think foundational detection still works, because GuardDuty receives independent streams.
  • If the question says scan new S3 objects for malware, think Malware Protection for S3, because S3 Protection focuses on data activity rather than object scanning.
  • If S3 malware scanning cannot access encrypted uploads, check the service role’s EventBridge, S3, and KMS permissions.

Remember

  • Multi-account membership can use AWS Organizations or accepted GuardDuty administrator invitations.

  • First-time foundational detection has a 30-day trial per Region.

  • Finding severity is 1.0 to 10.0; higher numbers indicate greater risk.

  • Malware Protection for S3 can be used without enabling GuardDuty.

  • Findings are retained for 90 days. New active findings are exported to EventBridge within about 5 minutes.

Trap: GuardDuty’s access to independent log streams does not provide retained customer logs. Configure the relevant logging service when you need to access or retain those logs.

Da documentação da AWS (19)
  • GuardDuty analyzes AWS data sources using threat intelligence and machine learning to identify suspicious and potentially malicious activity.
  • GuardDuty can detect compromised credentials and unauthorized cryptomining on EC2 instances and container workloads.
  • A GuardDuty finding represents a potential security issue detected in AWS accounts, workloads, or data.
  • GuardDuty consumes CloudTrail management events through an independent duplicate stream without changing existing CloudTrail configurations.
  • GuardDuty uses an independent duplicate stream of EC2 VPC flow logs without requiring the customer to create or configure flow logs.
  • GuardDuty extracts fields for profiling and anomaly detection and discards the foundational source logs.
  • S3 Protection analyzes S3 data events to detect potential risks such as data exfiltration and destruction.
  • EKS Audit Log Monitoring analyzes Kubernetes audit logs for suspicious activity.
  • Malware Protection for EC2 scans associated EBS volumes, while Malware Protection for S3 scans newly uploaded S3 objects.
  • Malware Protection for S3 can be used without enabling GuardDuty, unlike the other protection plans.
  • First-time GuardDuty use in a Region provides a 30-day free trial for foundational detection.
  • Finding severity ranges from 1.0 to 10.0, with higher values indicating greater risk.
  • Runtime Monitoring is not enabled by default when GuardDuty is enabled.
  • GuardDuty retains findings for 90 days and exports active findings to EventBridge; new active findings are exported within about 5 minutes.
  • S3 findings export requires a bucket policy allowing GuardDuty to upload and a KMS key policy allowing GuardDuty to generate a data key; source account and detector ARN can restrict access.
  • GuardDuty replicates relevant CloudTrail global service events for processing in each Region where GuardDuty is enabled.
  • The GuardDuty delegated administrator and members must be configured in each desired Region, using the same delegated administrator account across Regions.
  • Malware Protection for S3 requires a service role with permissions for the managed EventBridge rule and S3 objects; KMS access is needed for supported KMS-encrypted objects.
  • Amazon GuardDuty administrator accounts can manage members added through AWS Organizations or members that accept an administrator invitation.

Organization enablement and response

GuardDuty can be managed across an organization and can send findingsA finding is a report of a potential security issue detected in an AWS account, workload, or data. to automation. The delegated administratorA delegated administrator is an account designated to manage GuardDuty for associated member accounts. is the account chosen to manage GuardDuty for associated member accounts.

How it works

  1. The Organizations management account designates a delegated GuardDuty administrator. Repeat the setup in each desired Region, using the same administrator account. Designation enables GuardDuty for that administrator in the current Region only.
  2. Set auto-enable for GuardDuty and supported protection plans in each Region. Malware Protection for S3 is not included in organization auto-enable.
Setting Effect
ALL Enable for existing and new organization accounts
NEW Enable only for new accounts when they join
NONE Do not automatically enable new accounts; existing enablement remains
  1. GuardDuty publishes finding events to EventBridge. A rule matches source: aws.guardduty and detail-type: GuardDuty Finding, optionally filtering finding details, and sends matches to a targetA target is a resource or endpoint that receives events matching an EventBridge rule. such as Lambda or an SNS topic. Lambda can run response code; SNS can deliver notifications.
  2. Applicable member findings reach EventBridge in both the member account and the delegated administrator account. Use the finding’s accountId to identify the originating account. Generate sample findingsA sample finding uses placeholder values to help test and understand GuardDuty findings. with placeholder values to test rules and automation.
  3. To tailor detection, the administrator manages trusted and threat listsA threat list identifies known malicious sources that GuardDuty uses to generate findings. for members. Trusted entries suppress findings; threat entries identify known malicious sources. After changing a list in S3, activate it again and wait for Active status.

Entity lists can contain IP addresses and domains; legacy IP lists contain IP addresses only. SHA-256 file hashes, which identify file contents, are supported only in threat entity lists. A trusted entry takes precedence if it is also on a threat list.

For infrastructure deployments, an AWS CloudFormation custom resource can invoke Lambda to check GuardDuty state before enabling it. This avoids duplicate enablement failures. A suspended GuardDuty keeps its detector, so the example reports AlreadyEnabled and leaves monitoring off; a disabled GuardDuty has no detector, so the example creates a new one without the old findings.

Exam cues

  • If the question says cover existing and future accounts, think ALL, because NEW covers only accounts joining after the setting is applied.
  • If the question says an existing member opted into a Region after joining, think ALL or individual enablement, because NEW requires opting in before joining.
  • If the question says centralize member alerts, think an administrator-account EventBridge rule, because it can match applicable member findings.
  • If the question says quarantine a potentially compromised EC2 instance, think EventBridge and response code, because a finding alone is not an isolation action.
  • If the question says a changed S3 threat list has no effect, think reactivation, because GuardDuty uses the updated list only when it is Active.
  • If the question says test an alert without a real threat, think sample findings, because they contain placeholder values and can exercise the rule.
  • If the question says deploy GuardDuty where it may already exist, think a Lambda-backed CloudFormation custom resource, because it can check state before enabling.

Remember

  • New finding IDs reach EventBridge in near real time. Updates for recurring findings default to 6 hours; 15 minutes and 1 hour are alternatives.
  • GuardDuty retains findings for 90 days. Export to S3 for longer retention.
  • Automatically suppressed findings are not sent to EventBridge; manually archived findings continue to be sent.
  • Changing an EC2 security group does not terminate existing tracked connections.
  • In an opt-in Region, an administrator that has opted out cannot auto-enable GuardDuty for disabled members.

Trap: A GuardDuty finding does not itself quarantine an instance. A separate response must perform isolation; replacing security groups does not end existing tracked connections.

Route a finding to an automated response
12GuardDutyfindingEventBridgematching ruleLambdaresponse code
  1. GuardDuty publishes a finding event.

  2. A matching EventBridge rule invokes the response target.

Da documentação da AWS (30)
  • The Organizations management account can designate a delegated GuardDuty administrator to manage member accounts.
  • GuardDuty administration and member association are Regional, and the same delegated administrator account must be used across Regions.
  • Designating the delegated administrator automatically enables GuardDuty for that administrator only in the current Region.
  • The ALL auto-enable setting enables GuardDuty for existing and new organization accounts.
  • The NEW auto-enable setting enables GuardDuty automatically only for new member accounts joining the organization.
  • Changing auto-enable to NONE does not disable GuardDuty for existing accounts.
  • For NEW auto-enable in an opt-in Region, the member must opt in before joining the organization; ALL can cover an existing member.
  • If the delegated administrator opts out of an opt-in Region, auto-enable cannot enable GuardDuty for members where it is disabled.
  • Organization auto-enable preferences can be set for protection plans except Malware Protection for S3.
  • GuardDuty automatically publishes findings as events to EventBridge, including new and aggregated findings.
  • GuardDuty finding events use source aws.guardduty and detail-type GuardDuty Finding.
  • An EventBridge rule selects events and sends matching events to a target such as Lambda or an SNS topic.
  • EventBridge rules in the GuardDuty administrator account can act on applicable member findings as well as administrator findings.
  • GuardDuty publishes applicable member findings to EventBridge in the member account and makes them available to EventBridge rules in the administrator account.
  • The accountId field in finding details identifies the member account that generated the finding.
  • New finding IDs are sent in near real time; updates for subsequent occurrences default to six hours and can be set to 15 minutes or one hour.
  • GuardDuty stores findings for 90 days; findings can be exported to S3 for longer retention.
  • Findings automatically archived by suppression rules are not sent to EventBridge; manually archived findings continue to be sent.
  • Sample findings contain placeholder values and can test EventBridge events, filters, and automation.
  • The console generates a sample of each supported finding type, while the API or CLI can generate selected finding types.
  • AWS recommends isolating a potentially compromised EC2 instance with a dedicated security group and removing its other security groups.
  • Changing security groups does not terminate existing tracked connections on an EC2 instance.
  • For potentially compromised credentials, investigate the IAM entity and API call in the finding and determine whether the activity was legitimate.
  • Trusted lists suppress findings for trusted entries, while threat lists generate findings for activity involving known malicious entries.
  • In a multi-account environment only the GuardDuty administrator manages lists, which apply to member accounts.
  • GuardDuty uses a list only when its status is Active, and the list must be activated again after its S3 contents are updated.
  • A trusted entry takes precedence if it also appears in a threat list.
  • Entity lists support IP addresses and domains, while legacy IP lists support IP addresses only; SHA-256 file hashes are supported only in threat entity lists.
  • An AWS CloudFormation custom resource backed by Lambda can check GuardDuty state before enabling it to avoid a failed duplicate deployment.
  • In the AWS conditional enablement pattern, a suspended GuardDuty keeps its detector, so the pattern reports AlreadyEnabled and does not resume monitoring; a disabled GuardDuty has no detector, so the pattern creates a new one without the old findings.

Monitoring and finding delivery

GuardDuty monitoring and findingA finding is a report of a potential security issue detected in an AWS account, workload, or data. delivery depend on the data source, protection plan, and Regional configuration. Understand which signals GuardDuty receives and how recurring findings are reported before designing a centralized workflow.

How it works

  1. GuardDuty accesses EC2 DNS queries through AWS DNS resolvers without requiring Resolver query logging. It also receives an independent stream of EC2 VPC flow logsA VPC flow log captures information about IP traffic to and from network interfaces., without a customer-configured flow log.
  2. Select the protection plan for the signal: Runtime Monitoring analyzes operating system events on EKS, EC2, and ECS, including Fargate; RDS Protection analyzes supported database logins; Lambda Protection monitors Lambda network activity. The GuardDuty security agent provides EC2 runtime events.
  3. Configure the same delegated administratorA delegated administrator is an account designated to manage GuardDuty for associated member accounts. and its members in each desired Region. Choose organization auto-enable as ALL, NEW, or NONE. Malware Protection for S3 is excluded from these preferences.
  4. GuardDuty aggregates related recurring activity into the original finding, increasing its count and updating details. Earlier details can be replaced. Configure VPC Flow Logs to retain individual activity.
  5. Configure findings export per Region. An administrator’s export includes associated member findings generated in the same Region. The destination S3 bucket and KMS key must share a Region; opt-in destination Regions have additional cross-Region restrictions.

Exam cues

  • If EC2 uses a non-AWS DNS resolver, think the foundational DNS source cannot see those queries, because it accesses queries through AWS DNS resolvers.
  • If a team needs operating system events, think Runtime Monitoring and the security agent for EC2, because the agent supplies its runtime events.
  • If a team needs database login and Lambda network detection, think RDS Protection and Lambda Protection, respectively, because they analyze those signals.
  • If existing and future organization accounts need GuardDuty, think ALL in each desired Region, because auto-enable is Regional. Malware Protection for S3 is excluded.
  • If repeated activity appears in one finding, think aggregation, because GuardDuty updates its count and latest details. Retain VPC Flow Logs for individual activity.
  • If recurring findings need faster EventBridge notifications, change the update frequency, because new findings and updates have different timing.

Remember

  • New findings reach EventBridge near real time. Updates default to 6 hours and can be set to 15 minutes or 1 hour.
  • Changing organization auto-enable to NONE does not disable existing accounts.
  • Archived findings, including new occurrences of suppressed findings, are not exported by default.

Trap: A suppression rule can archive a recurring finding, so its new occurrences are not exported to S3 by default.

Da documentação da AWS (12)
  • GuardDuty accesses EC2 DNS queries through AWS DNS resolvers, not other DNS resolvers, and does not require Resolver query logging to be configured.
  • Runtime Monitoring analyzes operating system events on EKS, EC2, and ECS including Fargate.
  • The GuardDuty security agent provides runtime events for Runtime Monitoring on EC2.
  • RDS Protection analyzes supported database login activity, while Lambda Protection monitors Lambda network activity.
  • Archived findings, including new occurrences of suppressed findings, are not exported by default.
  • Findings export settings are Regional. The S3 bucket and KMS key must share a Region, and opt-in Region destinations have additional cross-Region restrictions.
  • A GuardDuty administrator's export includes findings from associated members generated in the same Region.
  • GuardDuty aggregates related recurring activity into the original finding, updating its count and latest details; earlier details can be replaced.
  • GuardDuty uses an independent duplicate stream of EC2 VPC flow logs without requiring the customer to create or configure flow logs.
  • New GuardDuty findings are sent to EventBridge near real time; updates to existing findings default to 6 hours and can be set to 15 minutes or 1 hour.
  • GuardDuty organization auto-enable supports ALL, NEW, and NONE; changing to NONE does not disable existing accounts. Malware Protection for S3 is excluded from these preferences.
  • The GuardDuty delegated administrator and members must be configured in each desired Region, using the same delegated administrator account across Regions.

Flashcards

toque para virar
Resposta

·

Todos os 34 flashcards em lista

A company wants continuous detection of compromised credentials and malicious activity in its AWS accounts without installing agents on its own. Which service fits?

Amazon GuardDuty.

Is Amazon GuardDuty a vulnerability scanner for software on EC2 instances?

No. GuardDuty detects threats from activity and logs; Amazon Inspector scans for software vulnerabilities.

Which AWS service detects potentially compromised credentials and unauthorized cryptomining by analyzing AWS activity?

GuardDuty. It analyzes foundational CloudTrail management events, EC2 VPC flow logs, and DNS logs.

A team has not configured a CloudTrail trail or VPC Flow Logs delivery. Can GuardDuty analyze foundational events?

Yes. GuardDuty consumes independent streams without changing the customer's CloudTrail or flow log configuration.

Which GuardDuty protection plans address suspicious S3 data activity and malware in newly uploaded S3 objects?

S3 Protection analyzes data events; Malware Protection for S3 scans newly uploaded objects.

A security team needs to detect potentially compromised AWS credentials and unauthorized cryptomining. Which service fits?

Amazon GuardDuty analyzes AWS data sources to identify suspicious activity, including compromised credentials and unauthorized cryptomining.

A security team needs GuardDuty findings retained beyond the service retention period. What should it configure?

Export findings to S3. GuardDuty retains findings for 90 days and sends active findings to EventBridge.

A central S3 bucket receives GuardDuty findings encrypted with a KMS key. What resource permissions are required?

Allow GuardDuty to upload in the bucket policy and generate a data key in the KMS key policy. Restrict access with the source account and detector ARN.

A company enables GuardDuty in one Region and expects centralized protection across every Region in AWS Organizations. What else must it configure?

Configure the same delegated administrator and its members in each desired Region. GuardDuty processes relevant global service events in each Region where it is enabled.

A team wants to scan newly uploaded Amazon S3 objects for malware. Which GuardDuty feature fits?

GuardDuty Malware Protection for S3 scans newly uploaded S3 objects.

An Amazon GuardDuty administrator needs a member relationship with another account. Must membership use AWS Organizations?

No. Members can join through AWS Organizations or by accepting the GuardDuty administrator account's invitation.

Does a GuardDuty finding automatically isolate a potentially compromised EC2 instance?

No. Use a response workflow, such as an EventBridge rule invoking Lambda, and configure its isolation action.

How can an organization enable GuardDuty for existing and new member accounts in multiple Regions?

Designate the same delegated administrator in each Region and select ALL for organization auto-enable in each Region.

How can an EventBridge rule route GuardDuty findings to an automated response?

Match source aws.guardduty and detail-type GuardDuty Finding, then send matching events to a target such as Lambda.

A GuardDuty finding indicates a potentially compromised EC2 instance. How can a team automate isolation?

Route the finding through EventBridge to Lambda response code that associates an isolation security group and removes the other groups. Existing tracked connections remain.

A GuardDuty finding suggests AWS credentials are compromised. What should the security team investigate?

Identify the IAM entity and API call in the finding, then determine whether the activity was legitimate.

What happens when GuardDuty organization auto-enable is changed from ALL to NONE?

Existing accounts are not disabled; NONE stops automatic enablement for new accounts.

An existing organization member opts into a Region after joining. Why might GuardDuty NEW auto-enable not work?

NEW requires the member to opt in before joining. Use ALL or manage the existing member individually in that Region.

How can a security team route GuardDuty member findings centrally and identify the originating account?

Use an EventBridge rule in the delegated administrator account and inspect the finding's accountId. Applicable findings also reach EventBridge in the member account.

A GuardDuty finding recurs after its first notification. What is the default EventBridge update frequency?

Six hours for subsequent occurrences. New finding IDs are sent in near real time.

How can a team test a GuardDuty EventBridge rule without generating a real threat?

Generate sample findings with placeholder values. The API or CLI can select specific finding types.

How do GuardDuty trusted and threat lists affect detection?

Trusted entries suppress findings; known malicious entries in threat lists can generate findings. Trusted entries take precedence when both lists contain the same entry.

A team edits a GuardDuty threat list stored in S3. What must it do for GuardDuty to use the changes?

Activate the list again and wait for its status to become Active.

Who manages GuardDuty trusted and threat lists for member accounts in a multi-account environment?

The GuardDuty administrator manages the lists, which apply to member accounts.

An AWS CloudFormation deployment may target accounts where GuardDuty already exists. How can it avoid a duplicate enablement failure?

Use a Lambda-backed custom resource to check GuardDuty state before enabling it. A suspended detector stays suspended (the pattern reports AlreadyEnabled); a disabled one is recreated without its old findings.

A GuardDuty finding is automatically archived by a suppression rule. Will EventBridge receive it?

No. Automatically suppressed findings are not sent to EventBridge; manually archived findings continue to be sent.

An EC2 instance uses a non-AWS DNS resolver. Can GuardDuty analyze its DNS queries through its foundational DNS source?

No. GuardDuty accesses that source through AWS DNS resolvers.

What does GuardDuty Runtime Monitoring analyze, and does it use an agent on EC2?

Operating system events on supported EKS, EC2, and ECS workloads, including Fargate. EC2 runtime events come from the GuardDuty security agent.

A team needs GuardDuty detection of unusual database logins and Lambda network activity. Which plans apply?

RDS Protection analyzes supported database logins; Lambda Protection analyzes Lambda network activity.

A GuardDuty suppression rule archives a recurring finding. Will new occurrences be exported to the configured S3 bucket by default?

No. Archived findings and new occurrences of suppressed findings are not exported by default.

A company exports GuardDuty member findings from multiple Regions to a central S3 bucket. What Regional configuration and encryption constraints apply?

Configure export in each source Region. The destination bucket and KMS key must share a Region; opt-in destination Regions have additional restrictions. An administrator exports associated member findings from its Region.

Repeated SSH brute-force attempts produce one GuardDuty finding instead of a new finding each time. How does GuardDuty handle the repeated activity?

It aggregates related occurrences, increasing the count and updating details. Older details can be replaced, so retain VPC Flow Logs for individual activity.

A team needs faster EventBridge notifications for recurring GuardDuty findings. Which setting should it change?

Set the update frequency to 15 minutes or 1 hour instead of the default 6 hours. New findings are sent near real time.

An organization wants GuardDuty automatically enabled for existing and future member accounts. Which auto-enable setting applies, and does it cover Malware Protection for S3?

Choose ALL in each desired Region. Malware Protection for S3 is not covered by organization auto-enable preferences.

Você concluiu todos os tópicos da DOP-C02!

Seu badge ficou dourado. Próximo passo: agendar a prova na AWS. Depois de passar, adicione-se ao Quem passou.

Compartilhar

Conte que você concluiu esta certificação

Concluí todos os tópicos da AWS Certified DevOps Engineer - Professional (DOP-C02) no Easy Review. Venha estudar também!