INFRASTRUCTURE LEARNINGINFRASTRUCTURE LEARNING
インフラ大全インフラ大全IT INFRASTRUCTURE / LEARNING
Infrastructure learning guide
Find a lesson or lab/

Esc to close · Search terms stay on your device

LEARNING日本語

Topic lessonTopic: Audit coverage and recovery design

CloudTrail event selection for S3 object activity

Decide which S3 object activity to capture before you need to investigate it. Use a fixed audit scenario to see precisely which records the chosen scope can provide.

What you will learn

Distinguish Event history from trail records and match bucket, prefix, operation and collection period to the audit question.

Prerequisite

This lesson builds on CloudTrail basics. All records and configurations are fictional.

Turn the audit question into a resource and an operation

CloudTrail Event history shows the last 90 days of management events in the selected Region. To investigate S3 data events, include them in a separate collection such as a trail.

From audit question to an interpretable result
From audit question to an interpretable result1: Audit question — Name the operation — Scope. 2: Scope — Record scope and time — Collection. 3: Collection — Check coverage — Empty results123Audit questionScopeCollectionEmpty resultsFrom audit question to an interpretable result1: Audit question — Name the operation — Scope. 2: Scope — Record scope and time — Collection. 3: Collection — Check coverage — Empty results123Audit questionScopeCollectionEmpty results
  1. Audit question → ScopeName the operation
  2. Scope → CollectionRecord scope and time
  3. Collection → Empty resultsCheck coverage
Audit question
Who submitted the data?
Scope
PutObject in the submission prefix
Collection
The trail with this selector
Empty results
Check time and scope

Decide which uploads and reads are observable

The scenario trail starts at 09:00 and captures only PutObject in the submission prefix
ActivityEvidence to examineCovered?
08:58 PutObject in submission prefixTrailBefore collection began
09:06 PutObject in submission prefixTrailIn scope; confirm delivery
09:08 GetObject from submission prefixTrailOperation outside scope
09:11 Bucket settings changedRegional Event historyCheck management events

Data-event selectors determine which resources and conditions are covered. An empty result cannot establish whether an out-of-scope operation happened.

Check coverage before drawing an audit conclusion

Write down the complete scope. Record the object key range, reads versus writes, Region and event time as well as the bucket.

Check effective time and delivery separately. Separate actions before collection, actions excluded by the selector, records still awaiting delivery, and searches using mismatched filters.

Keep evidence proportional. Record event IDs and the matching conditions. Limit access to event bodies that may contain user information.

Check your decision

Choose an option to see the result and explanation.

Where should you look for the 09:06 upload?
Read the answer and reason

Answer: The selected trail · Both the prefix and operation match the trail selector.

No GetObject record appears. What can you conclude?
Read the answer and reason

Answer: Reads remain unverified · The trail does not collect reads in this scenario.

Why can the 08:58 upload be absent?
Read the answer and reason

Answer: Collection had not started · The action predates the 09:00 start.

Check, then explain in your own words

  1. Write the audit question in one sentence.
  2. Record the operation, resource and time with the result.
  3. State what the evidence cannot establish.
Check your understanding: Can a write-only trail show both writes and reads?

Reads need another selector or another evidence source.

Audit conclusions are limited to the coverage of the records.

Sources and conditions checked

The official references used in this lesson are collected here with their scope, conditions and corresponding sections. Independent expert review and live AWS comparison have not been performed.