Credential Access · Term

What is LLMNR poisoning?

LLMNR poisoning lets an attacker on the local network answer name-resolution requests and trick Windows machines into sending their password hashes. It often needs no credentials at all. Here is how.

Credential Access · TermAll services
TL;DR

LLMNR poisoning is a local-network attack where an attacker answers LLMNR and NBT-NS name-resolution broadcasts that Windows sends when DNS fails, posing as the requested host. The victim then tries to authenticate to the attacker, sending its NetNTLM hash, which the attacker captures and cracks offline (or relays). Because Windows broadcasts these requests by default, the attack often needs no credentials, just a foothold on the network. The tool Responder automates it. Disabling LLMNR/NBT-NS is the fix.

By John Dill, Red Team Lead, SecureLayer7Updated

What LLMNR poisoning is

When a Windows machine cannot resolve a name through DNS (a typo, a stale share path), it broadcasts a request using LLMNR and NBT-NS, essentially asking the whole local subnet "who is this host?".

LLMNR poisoning is an attacker on that subnet answering "that’s me". The victim, believing it found the host, tries to authenticate, and in doing so sends its NetNTLM challenge-response, which the attacker collects. No DNS compromise, no credentials, just listening and replying.

The attack and payload

On a foothold in the local network, the attacker listens and answers:

  • Run a poisoner: responder -I eth0 (answers LLMNR/NBT-NS, captures NetNTLMv2 hashes).
  • Victims trying to reach mistyped or stale hosts hand over their NetNTLMv2 hashes.
  • Crack them offline: hashcat -m 5600 netntlm.txt rockyou.txt, or relay them to another host (see NTLM relay) without cracking.

Documented for defensive context.

How to defend

  • Disable LLMNR and NBT-NS via Group Policy and DHCP options, the direct fix; ensure DNS is correct so they are not needed.
  • Enforce SMB signing so captured hashes cannot be relayed.
  • Use strong passwords so any captured NetNTLMv2 hashes resist cracking.
  • Segment the network to limit where a poisoner can listen.
  • Monitor for LLMNR/NBT-NS responders and unusual authentication patterns.

References

  1. [1]MITRE ATT&CK: Adversary-in-the-Middle (T1557)(MITRE)
  2. [2]Microsoft: Windows name resolution(Microsoft)
  3. [3]MITRE ATT&CK: Credential Access (TA0006)(MITRE)
Related terms

Common questions

Credential access, asked often

Want your environment tested for exposed credentials?

Scope an engagement

Find the exposed credentials before an attacker does.

Our internal and network penetration tests hunt the credentials an intruder would, in memory, registry hives, config files, and on the wire, then show your team exactly where each one was exposed and how to close it. Free re-test included.

See all services30-min scoping call, fixed-price proposal in 48 hours.