Chapter 45 Quiz: Active Directory Red Teaming¶
Test your knowledge of BloodHound analysis, ACL abuse, Kerberoasting, delegation attacks, Active Directory Certificate Services (ADCS) exploitation, and golden/silver ticket attacks.
Questions¶
1. A red team operator runs BloodHound and identifies a path from a compromised standard user to Domain Admin through three intermediate group memberships and one ACL edge. What does this attack path represent?
- A) A vulnerability in BloodHound's graphing algorithm
- B) A chained privilege escalation path where each hop grants incrementally higher permissions, ultimately reaching Domain Admin through transitive relationships
- C) A misconfiguration in DNS that allows lateral movement
- D) A false positive caused by stale Active Directory objects
Answer
B — A chained privilege escalation path where each hop grants incrementally higher permissions, ultimately reaching Domain Admin through transitive relationships
BloodHound maps Active Directory relationships (group memberships, ACLs, sessions, GPO links) and identifies attack paths through graph theory. Multi-hop paths reveal how seemingly low-risk permissions chain together to create critical escalation routes. A path from a standard user to Domain Admin through transitive group memberships and ACL edges represents a real, exploitable privilege escalation chain.
2. An attacker discovers that their compromised user account has GenericAll permissions on another user object in Active Directory. What attack can the attacker perform?
- A) Only read the target user's attributes
- B) Perform any operation on the target object including password reset, SPN modification (for targeted Kerberoasting), or adding the account to privileged groups
- C) Only modify the target user's group memberships
- D) Only view the target user's Kerberos tickets
Answer
B — Perform any operation on the target object including password reset, SPN modification (for targeted Kerberoasting), or adding the account to privileged groups
GenericAll grants full control over the target object, equivalent to object ownership. On a user object, this enables: resetting the password without knowing the current one, modifying the servicePrincipalName attribute to enable targeted Kerberoasting, adding the user to groups, modifying logon scripts, and writing to any attribute. This is one of the most powerful ACL edges in BloodHound attack paths.
3. What is the fundamental difference between Kerberoasting and AS-REP Roasting in terms of prerequisites and target accounts?
- A) Kerberoasting targets machine accounts; AS-REP Roasting targets user accounts
- B) Kerberoasting requires any authenticated domain account and targets accounts with SPNs; AS-REP Roasting targets accounts with Kerberos pre-authentication disabled and can be performed without domain credentials
- C) Kerberoasting requires domain admin; AS-REP Roasting requires only standard user access
- D) Kerberoasting works against Linux systems; AS-REP Roasting works against Windows systems
Answer
B — Kerberoasting requires any authenticated domain account and targets accounts with SPNs; AS-REP Roasting targets accounts with Kerberos pre-authentication disabled and can be performed without domain credentials
Kerberoasting requests TGS tickets for service accounts with registered SPNs — the returned ticket is encrypted with the service account's hash and can be cracked offline. AS-REP Roasting targets accounts where the DONT_REQUIRE_PREAUTH flag is set, requesting an AS-REP that contains encrypted material crackable offline. AS-REP Roasting can be performed with just a list of usernames and no domain credentials.
4. A red team operator performs a DCSync attack using Mimikatz (lsadump::dcsync /domain:corp.local /user:krbtgt). What Active Directory permission is required for this attack, and what does it retrieve?
- A) Local administrator on the domain controller; retrieves the SAM database
- B) Replicating Directory Changes and Replicating Directory Changes All rights; retrieves the krbtgt account's NTLM hash by simulating domain controller replication
- C) Schema Admin privileges; retrieves the Active Directory schema
- D) DNS Admin privileges; retrieves DNS zone transfer data
Answer
B — Replicating Directory Changes and Replicating Directory Changes All rights; retrieves the krbtgt account's NTLM hash by simulating domain controller replication
DCSync abuses the Directory Replication Service (DRS) protocol to request password data as if the attacker were a domain controller performing replication. It requires Replicating Directory Changes and Replicating Directory Changes All permissions, which Domain Admins and domain controllers possess by default. Retrieving the krbtgt hash enables Golden Ticket creation for unlimited domain persistence.
5. What is the difference between a Golden Ticket and a Silver Ticket in terms of scope, creation requirements, and detection difficulty?
- A) Golden Tickets are created with the domain admin hash; Silver Tickets with any user hash
- B) Golden Tickets are forged TGTs created with the krbtgt hash, granting access to any service in the domain; Silver Tickets are forged TGS tickets created with a specific service account hash, granting access only to that service but without contacting the KDC
- C) Golden Tickets work across forests; Silver Tickets work within a single OU
- D) Golden Tickets expire after 10 hours; Silver Tickets never expire
Answer
B — Golden Tickets are forged TGTs created with the krbtgt hash, granting access to any service in the domain; Silver Tickets are forged TGS tickets created with a specific service account hash, granting access only to that service but without contacting the KDC
Golden Tickets forge TGTs signed with the krbtgt hash, allowing the attacker to request TGS tickets for any service as any user (including non-existent ones). Silver Tickets forge TGS tickets for a specific service using that service's NTLM hash. Silver Tickets are harder to detect because they never touch the KDC — the forged TGS is presented directly to the target service.
6. An attacker discovers that a compromised account has WriteDacl permissions on the domain object. How can this be abused for privilege escalation?
- A) The attacker can read all domain objects but cannot modify them
- B) The attacker can modify the domain object's DACL to grant themselves DCSync rights (Replicating Directory Changes), then extract all domain credentials
- C) The attacker can only modify password policies
- D) The attacker can delete the domain object
Answer
B — The attacker can modify the domain object's DACL to grant themselves DCSync rights (Replicating Directory Changes), then extract all domain credentials
WriteDacl allows modification of the object's Discretionary Access Control List. On the domain object, this means the attacker can add an ACE granting their compromised account the Replicating Directory Changes and Replicating Directory Changes All extended rights, enabling a DCSync attack. This is a critical ACL abuse path that BloodHound specifically identifies and highlights.
7. What is Constrained Delegation, and how can it be abused by an attacker who compromises an account configured with this setting?
- A) Constrained Delegation limits which users can log on to a server
- B) Constrained Delegation allows a service to impersonate users to specific backend services; an attacker can abuse it using S4U2Self and S4U2Proxy to obtain service tickets for the allowed services as any domain user
- C) Constrained Delegation restricts which GPOs apply to a computer
- D) Constrained Delegation limits Kerberos ticket lifetime
Answer
B — Constrained Delegation allows a service to impersonate users to specific backend services; an attacker can abuse it using S4U2Self and S4U2Proxy to obtain service tickets for the allowed services as any domain user
Constrained Delegation (configured via msDS-AllowedToDelegateTo) permits a service to request tickets on behalf of users for specific SPNs. An attacker with the service's credentials can use S4U2Self to obtain a forwardable ticket for any user, then S4U2Proxy to present that ticket to the allowed backend services. If the allowed SPN includes CIFS or HTTP on a domain controller, this can escalate to domain compromise.
8. A red team operator discovers an Active Directory Certificate Services (ADCS) server with a certificate template that allows low-privileged users to enroll and specify a Subject Alternative Name (SAN). Why is this a critical vulnerability?
- A) It allows denial of service against the CA
- B) The attacker can request a certificate with an arbitrary SAN (e.g., the Domain Admin UPN), then use that certificate for Kerberos authentication as the privileged user (ESC1)
- C) It exposes the CA's private key
- D) It allows the attacker to revoke all issued certificates
Answer
B — The attacker can request a certificate with an arbitrary SAN (e.g., the Domain Admin UPN), then use that certificate for Kerberos authentication as the privileged user (ESC1)
ESC1 (Escalation Scenario 1) in ADCS exploitation occurs when a certificate template allows enrollees to specify the Subject Alternative Name (SAN) and has enrollment rights for low-privileged users. The attacker enrolls with a SAN containing a Domain Admin's UPN, receives a valid certificate, and uses it for PKINIT Kerberos authentication. This effectively impersonates the Domain Admin with a CA-signed certificate.
9. What is Resource-Based Constrained Delegation (RBCD), and how does it differ from traditional Constrained Delegation in terms of configuration and attack prerequisites?
- A) RBCD requires Schema Admin to configure; traditional delegation requires Domain Admin
- B) RBCD is configured on the target (backend) service via the
msDS-AllowedToActOnBehalfOfOtherIdentityattribute, and can be abused by anyone with write access to that attribute — unlike traditional delegation, which requires Domain Admin to configure - C) RBCD only works in Azure AD; traditional delegation only works on-premises
- D) RBCD uses NTLM; traditional delegation uses Kerberos
Answer
B — RBCD is configured on the target (backend) service via the msDS-AllowedToActOnBehalfOfOtherIdentity attribute, and can be abused by anyone with write access to that attribute — unlike traditional delegation, which requires Domain Admin to configure
RBCD shifts delegation configuration from the front-end service (requiring SeEnableDelegationPrivilege, typically Domain Admin) to the back-end resource. An attacker with write access to a computer's msDS-AllowedToActOnBehalfOfOtherIdentity attribute can configure a controlled account to delegate to that computer, then use S4U2Self and S4U2Proxy to access the target as any user.
10. A red team operator discovers that Group Policy Preferences (GPP) files in SYSVOL contain encrypted passwords. Why are these passwords a security risk despite being encrypted?
- A) GPP uses strong encryption that requires quantum computing to break
- B) Microsoft published the AES-256 decryption key in MSDN documentation, making all GPP-encrypted passwords trivially decryptable using tools like gpp-decrypt
- C) GPP passwords are encrypted with the domain controller's machine account hash
- D) GPP passwords are only accessible to Domain Admins
Answer
B — Microsoft published the AES-256 decryption key in MSDN documentation, making all GPP-encrypted passwords trivially decryptable using tools like gpp-decrypt
Group Policy Preferences allowed administrators to push local admin passwords, drive mappings, and scheduled tasks via SYSVOL. The passwords were encrypted with a static AES-256 key that Microsoft published in MSDN documentation (MS14-025). Any authenticated domain user can read SYSVOL, find cpassword values in Groups.xml, ScheduledTasks.xml, and similar files, and instantly decrypt them.
11. An attacker has compromised an account with ForceChangePassword ACL rights on a target user. What can the attacker do, and how does this differ from GenericAll?
- A) The attacker can view the target's current password hash
- B) The attacker can reset the target's password to a known value without knowing the current password, but unlike GenericAll, cannot modify other attributes such as SPNs or group memberships
- C) The attacker can only unlock the target account
- D) The attacker can force the target to change their password at next logon
Answer
B — The attacker can reset the target's password to a known value without knowing the current password, but unlike GenericAll, cannot modify other attributes such as SPNs or group memberships
ForceChangePassword (User-Force-Change-Password extended right) allows resetting a user's password without knowing the current password. This is more limited than GenericAll, which grants full control over all object attributes. However, password reset alone is sufficient to take over the target account and is a viable privilege escalation step in multi-hop BloodHound attack paths.
12. What is the purpose of the AdminSDHolder mechanism in Active Directory, and how can it be abused for persistence?
- A) AdminSDHolder manages DNS records for domain controllers
- B) AdminSDHolder automatically propagates its ACL to all protected groups (Domain Admins, Enterprise Admins, etc.) every 60 minutes via SDProp; an attacker who modifies the AdminSDHolder ACL gains persistent access to all protected objects
- C) AdminSDHolder stores backup domain admin credentials
- D) AdminSDHolder controls Group Policy processing order
Answer
B — AdminSDHolder automatically propagates its ACL to all protected groups (Domain Admins, Enterprise Admins, etc.) every 60 minutes via SDProp; an attacker who modifies the AdminSDHolder ACL gains persistent access to all protected objects
The Security Descriptor Propagation (SDProp) process runs every 60 minutes and overwrites the ACL of protected accounts and groups with the AdminSDHolder container's ACL. An attacker with sufficient access can add an ACE to AdminSDHolder granting their controlled account permissions over all protected objects. Even if defenders clean up the compromised account's direct permissions, SDProp restores them within an hour.
13. During an AD red team engagement, the operator discovers a trust relationship between the current domain and a parent domain. What attack can exploit the SID History attribute to escalate from the child domain to the parent domain?
- A) Pass-the-Hash across the domain trust
- B) Create a Golden Ticket in the child domain with the Enterprise Admins SID from the parent domain injected into the SID History, enabling cross-domain escalation via SID History injection
- C) Kerberoast service accounts in the parent domain
- D) Perform LLMNR poisoning across the trust boundary
Answer
B — Create a Golden Ticket in the child domain with the Enterprise Admins SID from the parent domain injected into the SID History, enabling cross-domain escalation via SID History injection
In a parent-child trust relationship, an attacker who compromises the child domain's krbtgt hash can forge a Golden Ticket that includes the Enterprise Admins SID (from the parent domain) in the SID History field. When this ticket is presented across the trust boundary, the parent domain grants Enterprise Admin privileges because it trusts the child domain's krbtgt-signed ticket and honors the SID History.
14. A red team operator needs to enumerate all Kerberoastable accounts in the domain. Which LDAP filter would most effectively identify service accounts with SPNs?
- A)
(objectClass=computer) - B)
(&(objectCategory=person)(objectClass=user)(servicePrincipalName=*)(!(cn=krbtgt))) - C)
(memberOf=Domain Admins) - D)
(userAccountControl=514)
Answer
B — (&(objectCategory=person)(objectClass=user)(servicePrincipalName=*)(!( cn=krbtgt)))
This LDAP filter identifies user accounts (not computer accounts) that have at least one SPN registered, excluding the krbtgt account. These accounts are Kerberoastable — any authenticated user can request TGS tickets for their SPNs and attempt offline cracking. Tools like Rubeus, Impacket's GetUserSPNs, and PowerView automate this enumeration and ticket extraction process.
15. An Active Directory environment has implemented tiered administration (Tier 0 for domain controllers, Tier 1 for servers, Tier 2 for workstations). A red team operator compromises a Tier 1 server and finds a Tier 0 admin session. What does this indicate about the tiering implementation?
- A) The tiering model is working correctly — Tier 0 admins should manage Tier 1 servers
- B) Tier 0 credential exposure on a Tier 1 asset violates the tiered administration model, indicating that privileged access boundaries are not properly enforced
- C) The operator should escalate to Tier 2 first before attacking Tier 0
- D) Tier 0 sessions on Tier 1 servers are acceptable during maintenance windows
Answer
B — Tier 0 credential exposure on a Tier 1 asset violates the tiered administration model, indicating that privileged access boundaries are not properly enforced
The tiered administration model specifically prohibits higher-tier credentials from being used on lower-tier assets. A Tier 0 admin session on a Tier 1 server means those credentials can be extracted from memory (via Mimikatz or similar), breaking the entire tiering model. This is a critical finding in AD red team assessments, demonstrating that administrative practices undermine the architectural security controls.
Scoring¶
| Score | Performance |
|---|---|
| 14–15 | Expert — Active Directory red teaming concepts fully internalized |
| 11–13 | Proficient — Ready to operate in AD-focused red team engagements |
| 8–10 | Developing — Review Chapter 45 BloodHound, ACL abuse, and delegation sections |
| <8 | Foundational — Re-read Chapter 45 before proceeding |
Return to Chapter 45 | Next: Chapter 46 Quiz