CloudTrail Not Enabled in All Regions

No CloudTrail trail records management events in every AWS Region, so API activity in some Regions, and activity older than 90 days, is not kept for investigation.

SeverityMedium
CVSS 3.14.3 CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:L/A:N Open in the calculator
CVSS 4.05.3 CVSS:4.0/AV:N/AC:L/AT:N/PR:L/UI:N/VC:N/VI:L/VA:N/SC:N/SI:N/SA:N Open in the calculator
CWECWE-778: Insufficient Logging
OWASP Top 10A09:2025 Security Logging and Alerting Failures
MITRE ATT&CKT1535 Unused/Unsupported Cloud Regions
CategoryInsufficient Logging

Description

AWS CloudTrail records API calls made in an account. Without any configuration, the CloudTrail event history shows the last 90 days of management events in each Region, but it keeps nothing longer, does not include data events such as reads of S3 objects, and cannot be sent to a central log store or used to raise alerts. A durable record needs a trail, or a CloudTrail Lake event data store, delivering events to storage the organisation controls.

The finding is raised when there is no trail, or when trails cover only some Regions. A trail created through the API or CLI covers only its own Region unless IsMultiRegionTrail is set, and activity in other Regions then goes unrecorded. An attacker with access to the account can work in Regions without a trail, which they may choose precisely because nobody watches them, or rely on events ageing out of the 90-day event history, knowing that their actions will not be retained. Without a complete record, the organisation may be unable to establish what an intruder did, which data they reached, or whether they created persistence such as new users, access keys or roles, and may be unable to meet its obligations to investigate and report a breach. The weakness gives no access on its own; its effect is on detection and response once another weakness has been exploited.

Related weaknesses often found at the same time include trails that are not logging, event selectors that exclude management events, log file integrity validation being disabled, and a log bucket that the account's own administrators can alter or delete. The configuration is reviewed with aws cloudtrail describe-trails and aws cloudtrail get-trail-status in each Region, and is reported by Prowler, ScoutSuite and Security Hub, which check for a multi-Region trail recording management events.

Remediation

Create a multi-Region trail, or an organisation trail covering every account, and protect its logs:

aws cloudtrail create-trail --name org-trail --s3-bucket-name "$LOG_BUCKET" --is-multi-region-trail --enable-log-file-validation --kms-key-id "$KMS_KEY_ARN" aws cloudtrail start-logging --name org-trail
  • In an AWS organisation, create the trail from the management account or a delegated administrator with --is-organization-trail, so that it covers every member account and member accounts cannot change or delete it.
  • Deliver logs to a bucket in a dedicated log archive account, with a bucket policy that lets only CloudTrail write to it, S3 Versioning enabled and, where required, S3 Object Lock to prevent deletion.
  • Enable log file integrity validation, and encrypt logs with a KMS key whose policy restricts decryption to the people who investigate incidents.
  • Add data events for sensitive S3 buckets and other high-value resources, and send events to the organisation's SIEM or to CloudWatch Logs, with alerts on high-risk actions such as StopLogging, DeleteTrail and root user activity.

Confirm the fix with aws cloudtrail describe-trails showing IsMultiRegionTrail as true, and aws cloudtrail get-trail-status showing that the trail is logging and delivering recently.

References


Add This Finding to Your Library

In Pental this write-up is in Settings, Library. Add to my library copies it into your own, rated in your firm's CVSS version, ready to put in your own words.

Free trial  ·  No card required

Only need the reporting? Our reporting tool is free and open source, with the same library.


More AWS Findings

Every AWS finding