Drill Case
Cloud Identity Case
A cloud account acting out of character: read its CloudTrail records and sign-ins, name the attack, and pick the containment.
The case
A cloud account did something it has never done before. The AWS CloudTrail records and the identity provider's sign-in logs are all you have, and the account is still active.
First decide what the identity was actually doing in AWS. Then work back to how it got in, and pick the containment that closes the door rather than one that only looks busy.
- CloudTrail 12 questions
- Identity 10 questions
What you will be able to do
- Read a CloudTrail record and pull the caller, the target and the time into a timeline row (CloudTrail Event Reference)
- Tell apart the caller and the target of an IAM call, which is what turns a routine event into a finding (CloudTrail Event Reference)
- Place a CloudTrail event in the stage of an intrusion it represents, and order events into a timeline
- Avoid the three conclusions CloudTrail invites you to get wrong: absent data events, us-east-1 global services, and a logging gap (CloudTrail Event Reference)
- Tell password spray, brute force, MFA fatigue, token theft and consent phishing apart from log evidence (Entra ID Sign-in Reference)
- Choose the containment action that actually resolves a finding: reset, revoke sessions, or revoke a grant (M365 Audit Log Reference)
- Read a sign-in row for the fields that decide urgency, including on sign-ins that succeeded
- Recognize when the log cannot answer the question, and say so instead of guessing (M365 Audit Log Reference)
Read first: CloudTrail Event Reference, Entra ID Sign-in Reference, M365 Audit Log Reference.
Notes for teaching it
- The data-events question is the one worth stopping a class on: it is the difference between "the data was not read" and "we were not logging reads", and confidently reporting the first when only the second is true is a real and common failure.
- Every record here is fabricated but structurally faithful: real field names in real positions, the AWS documentation account number, RFC 5737 addresses. Learners can read them exactly as they would read production records.
- No AWS account or prior cloud experience is needed. The two ordering questions in particular test investigative reasoning rather than AWS knowledge, and they work well as a group discussion.
- The matching question on containment is the one to build a session around: the instinct that a password reset fixes everything resolves exactly one of its six findings, and watching that fail is more persuasive than being told.
- Consent phishing surprises almost everyone the first time, because it produces no failed sign-in, no password change, and a user who genuinely approved it. Worth pausing on.
- The last question has no satisfying answer by design. Saying "the log cannot tell us, and here is why" is the professional response, and learners need permission to give it.
Terms behind it
- Cloud Audit Logs: Records of management-plane API activity in a cloud environment.
- Valid Accounts: Logging in with legitimate stolen or default credentials.
- MFA Fatigue: Spamming MFA push prompts until the user approves one.
- Illicit OAuth Grant: Tricking a user into authorizing a malicious app's access to their account.
- Containment: Stopping an incident from spreading while you investigate.
- Temporary Credentials: Time-limited, automatically expiring credentials issued for a session or role.
Worksheet
Every question on one page, to print or to read without JavaScript. The answer key has the answers and why.
CloudTrail, 12 questions
-
This is the only ConsoleLogin record for this identity in the last 90 days. What is the most important thing about it?
{ "eventTime": "2026-03-02T02:14:51Z", "eventSource": "signin.amazonaws.com", "eventName": "ConsoleLogin", "awsRegion": "us-east-1", "sourceIPAddress": "198.51.100.23", "userAgent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64)", "userIdentity.type": "IAMUser", "userIdentity.arn": "arn:aws:iam::123456789012:user/svc-backup", "responseElements.ConsoleLogin": "Success", "additionalEventData.MFAUsed": "No" }- A service account signed in to the console interactively, and MFA was not used
- The sign-in came from outside the corporate IP range
- The user agent is a browser rather than the AWS CLI
- The event was recorded in us-east-1 rather than the account's home region
-
Pull the four fields you would put in a timeline row for this record.
{ "eventVersion": "1.09", "eventTime": "2026-03-02T02:31:44Z", "eventSource": "iam.amazonaws.com", "eventName": "CreateAccessKey", "awsRegion": "us-east-1", "sourceIPAddress": "198.51.100.23", "userAgent": "aws-cli/2.15.30 Python/3.11.8 Linux/6.5.0", "userIdentity.arn": "arn:aws:iam::123456789012:user/svc-backup", "requestParameters.userName": "admin-breakglass", "readOnly": "false" } -
One identity produced 34 records in four minutes. Thirty-one carry errorCode "AccessDenied", across 20 different eventName values. What is this?
- An identity mapping the edges of its own permissions: enumeration by trial and error
- A misconfigured application retrying a failing call
- Normal noise; AccessDenied is not recorded by CloudTrail
- A throttling condition that should be read as a performance issue
-
Sort each CloudTrail event into the stage of an intrusion it most directly represents.
-
These five events all came from one identity inside eleven minutes. Put them in the order that reconstructs the intrusion.
-
A bucket held customer data. You have the full CloudTrail export for the incident window and there is not a single GetObject record. What can you conclude?
- Nothing about whether the data was read, because S3 object-level events are not recorded unless data events were enabled
- The data was not read, because CloudTrail records every S3 API call
- The data was read but the records were deleted, since GetObject is always logged
- The bucket was empty during the incident window
-
The compromised workload runs entirely in eu-west-1. You pull CloudTrail for eu-west-1 only and see no IAM activity at all. Why is that not reassuring?
- IAM, STS and CloudFront are global services and record their events in us-east-1 regardless of caller location
- IAM events are only recorded in the account's billing region
- IAM activity is never written to CloudTrail; it goes to AWS Config
- Global service events are delayed by 24 hours and had not yet arrived
-
You need to follow what this session did next. Pull the three fields that let you pivot from this record to the rest of the activity.
{ "eventTime": "2026-03-02T02:12:03Z", "eventSource": "sts.amazonaws.com", "eventName": "AssumeRole", "awsRegion": "us-east-1", "sourceIPAddress": "198.51.100.23", "userIdentity.arn": "arn:aws:iam::123456789012:user/svc-backup", "requestParameters.roleArn": "arn:aws:iam::123456789012:role/OrgAdminRole", "requestParameters.roleSessionName": "ops-maintenance", "responseElements.assumedRoleUser.assumedRoleId": "AROAEXAMPLEID:ops-maintenance" } -
What has happened here, and why does it matter after you rotate every credential in the account?
{ "eventTime": "2026-03-02T02:44:10Z", "eventSource": "iam.amazonaws.com", "eventName": "UpdateAssumeRolePolicy", "userIdentity.arn": "arn:aws:iam::123456789012:user/svc-backup", "requestParameters.roleName": "OrgAdminRole", "requestParameters.policyDocument": "{\"Statement\":[{\"Effect\":\"Allow\",\"Principal\":{\"AWS\":\"arn:aws:iam::210987654321:root\"},\"Action\":\"sts:AssumeRole\"}]}" }- An outside AWS account can now assume an admin role here, and rotating credentials does not revoke that
- The role was deleted and recreated, so its permissions were reset
- A new IAM user was created in account 210987654321
- The role can no longer be assumed by anyone, which is a denial of service
-
StopLogging appears at 02:51:07Z and the trail resumes at 06:20:00Z when an engineer notices. What is the correct reading of the gap?
- The gap is itself evidence, and other sources may still cover it: a second trail, an Organization trail, or already-delivered S3 objects
- Everything in the gap is unrecoverable, so the investigation ends at 02:51:07Z
- StopLogging is not itself recorded, so the start of the gap cannot be established
- The gap means the trail was deleted and its prior records are gone
-
Order the events of an EBS snapshot exfiltration, from reconnaissance to the data leaving your control.
-
A quarterly access review confirmed that no new policies were attached to any identity. Two weeks earlier, CreatePolicyVersion ran with setAsDefault=true on a policy already attached to a service role. Why did the review miss the escalation?
- The attachment never changed; only the policy's contents did, and the review checked attachments rather than current permissions
- CreatePolicyVersion does not affect permissions until the policy is re-attached
- Policy versions are not recorded in CloudTrail
- The review would have caught it; the escalation must have come from somewhere else
Identity, 10 questions
-
Twelve minutes of sign-in failures from one source address. What is this, and what does it tell you to do next?
09:02:11 alice@example.com AADSTS50126 198.51.100.77 09:02:14 bob@example.com AADSTS50126 198.51.100.77 09:02:18 carol@example.com AADSTS50034 198.51.100.77 09:02:22 dan@example.com AADSTS50126 198.51.100.77 09:02:26 erin@example.com AADSTS50057 198.51.100.77 09:02:31 frank@example.com AADSTS50126 198.51.100.77 … 340 more, 331 distinct user principal names, 1-2 attempts each
- Password spray: one or two guesses against many accounts, so look for the successes rather than the failures
- Brute force against alice@example.com, so lock that one account
- A misconfigured mail client retrying, so contact the user
- Smart lockout working correctly, so no action is needed
-
What happened to this account, and what does the last line mean?
23:41:02 dan@example.com AADSTS500121 MFA 203.0.113.14 23:41:39 dan@example.com AADSTS500121 MFA 203.0.113.14 23:42:15 dan@example.com AADSTS500121 MFA 203.0.113.14 23:42:50 dan@example.com AADSTS500121 MFA 203.0.113.14 23:43:31 dan@example.com AADSTS500121 MFA 203.0.113.14 23:44:08 dan@example.com Success MFA 203.0.113.14
- MFA fatigue. The attacker already had the password, and the user eventually approved a prompt
- The user forgot their password five times and then remembered it
- A network problem caused five timeouts before the sign-in succeeded
- Smart lockout released the account after five failures
-
Match each finding to the containment action that actually resolves it. Getting these wrong is how an incident continues after it is declared closed.
-
Pull the four fields that decide how urgent this successful sign-in is.
Date (UTC) : 2026-03-02 04:12:56 User : erin@example.com Status : Success IP address : 203.0.113.14 Location : Lagos, NG Client app : IMAP4 Authentication requirement : singleFactorAuthentication Conditional Access : Not applied Risk state : At risk Risk detection : Unfamiliar sign-in properties
-
A user reports a suspicious "document share" email. There are no failed sign-ins for that account and no password has changed, but the audit log has this entry. What happened?
Operation : Consent to application User : frank@example.com App : "Secure Docs Viewer" (third-party) Scopes : Mail.Read, Mail.Send, Files.Read.All, offline_access Result : Success
- An illicit consent grant. The attacker holds a refreshing token and never needed the password at all
- A phishing attempt that failed, because no sign-in failure was recorded
- A routine software installation, since the user approved it themselves
- A password compromise that will be resolved by resetting the credential
-
Put this business email compromise in the order it happened.
-
Before the spray in question 1, the same source produced 900 sign-in attempts of which 780 returned AADSTS50034. What was that phase, and what did the attacker gain?
- User enumeration. The 120 accounts that did NOT return 50034 are the real ones, and became the spray target list
- A denial-of-service attempt against the tenant sign-in endpoint
- A misconfigured directory sync producing errors for deleted users
- Nothing useful; AADSTS50034 is returned for every failed sign-in
-
A sign-in from an unrecognized country returned AADSTS50074. MFA was required and not satisfied. The account is safe. What, if anything, still needs to happen?
- Reset the password. It was accepted before MFA was challenged, so it is known to the attacker
- Nothing; MFA blocked the sign-in, which is the control working as designed
- Disable the account permanently, since the location is untrusted
- Revoke the user's OAuth grants, since MFA failures imply token theft
-
What is the concern here, given that no user account was involved?
Operation : Add service principal credentials Actor : gina@example.com (Application Administrator) Target : "Contoso Reporting Connector" (existing app, created 2023) Change : new client secret added, expires in 24 months
- An already-trusted, already-permissioned application can now be authenticated as by whoever holds the new secret
- The application will stop working when the old secret expires
- Nothing: rotating a client secret is routine maintenance
- The Application Administrator role was escalated by this action
-
A mailbox compromise is confirmed. Leadership asks which messages were read so they can scope notification. The tenant is licensed E3. What is the honest answer?
- It cannot be determined from the audit log. MailItemsAccessed requires Purview Audit (Premium), so the record does not exist
- No messages were read, because there are no MailItemsAccessed records
- Every message in the mailbox was read, since the attacker had full access
- Only the messages matching the attacker's inbox rule were read
Next: Network Case. All drills