Separating Pass-the-Hash from Relay

Authentication material for one lab account may suggest a path to another machine, but the form of that material matters. Separating pass-the-hash from relay prevents two different mechanisms from collapsing into one vague idea of credential reuse. Is pass-the-hash actually the…

Authentication material for one lab account may suggest a path to another machine, but the form of that material matters. Separating pass-the-hash from relay prevents two different mechanisms from collapsing into one vague idea of credential reuse. Is pass-the-hash actually the appropriate next test? Compare the required material, the protocol conditions, and the evidence that would establish the resulting identity and access.

Would relay make more sense? Or would either attempt be based on assumptions you haven’t checked? The useful question isn’t whether you recognize a tool or command. It’s whether the material you possess, the protocol in use, the account’s privileges, and the destination’s configuration can combine to produce a meaningful change in access.

Begin by defining that change. Perhaps your current account can read files on one workstation, and you’re investigating whether the captured material permits authenticated access to a second host. Your baseline should identify the source and destination systems, the identity you currently control, the service you intend to test, and what you can do before the test.

Then state what success would look like. An authenticated connection is one possible observation, but it may not be enough. If your hypothesis concerns administrative access, you also need evidence that the resulting identity has the necessary privileges on that destination.

Windows credential work includes understanding NTLM and Net-NTLMv2 acquisition, offline cracking, pass-the-hash, and relay concepts in controlled labs. These concepts are connected, but they aren’t interchangeable. A captured NTLM password hash, a challenge-response exchange, and a recovered plaintext password are different forms of authentication material. Each supports different possible actions. Calling all of them a hash hides the distinction that should guide your next test.

Pass-the-hash means using a captured password hash as authentication material without first recovering the plaintext password. In the relevant Windows authentication path, possession of suitable hash material may let an operator authenticate as the associated account. That’s why the technique can support lateral movement between systems. It doesn’t mean the hash is a universal password, and it doesn’t mean possession guarantees useful access everywhere. The destination still has to accept the applicable authentication protocol, recognize or validate the account, and grant that account some useful authorization after authentication.

Relay is different. Instead of taking a stored password hash and presenting it later as reusable account material, a relay places itself between an authenticating client and a destination service. The client participates in a challenge-response authentication exchange.

The relay forwards the relevant messages to another service while that exchange is active, attempting to have the destination accept the client’s authentication. The relay doesn’t need to turn the captured exchange into the plaintext password. Its opportunity depends on the protocols, timing, service behavior, and protections involved in that specific path.

That distinction gives you an immediate decision point. If you possess an NTLM password hash associated with an account, pass-the-hash may be the hypothesis to evaluate. If you can induce or observe an eligible challenge-response exchange and have a reachable destination that may accept it, relay may be the relevant hypothesis.

A previously captured Net-NTLMv2 response isn’t automatically equivalent to the stored NTLM password hash needed for pass-the-hash. Nor should you assume that saving a challenge-response exchange means you can replay it whenever you choose. The challenge and protocol conversation matter.

Now consider a deliberately vulnerable lab with three machines. You control a low-privilege session on a workstation. A file server and an administrative server are reachable from it. Your notes contain NTLM hash material for a synthetic lab account named Lab Support.

Before testing anything, you establish that your current workstation identity can’t access a protected share on the file server. That failed access is useful as a baseline. You also confirm which destination is authorized for testing and which service is exposed. Without those details, a later success can’t be connected confidently to the authentication material.

Your prediction might be that the Lab Support hash will authenticate to the file server, but that prediction contains at least two separate claims. First, the destination will accept the authentication attempt as Lab Support. Second, Lab Support will be authorized to perform the action you care about. Authentication establishes an identity.

Authorization determines what that identity may do. A successful login followed by an access-denied response can therefore mean the supplied material was accepted but the account lacked permission for that resource. Treating that result as total failure would discard useful evidence. Treating it as administrative compromise would overstate it.

Account scope also matters. A domain account can be validated through the domain environment when the required path is available. A local account belongs to the individual Windows system.

Similar account names on two machines don’t, by themselves, prove that they represent the same security identity or use the same password material. Before interpreting a result, record whether you believe the material belongs to a local or domain account and why. If you haven’t validated that belief, mark it as a hypothesis rather than quietly turning it into a fact.

Protocol support is another gate. NTLM continues to exist alongside Kerberos in Windows environments, but that doesn’t mean every attempted connection will use NTLM. The service, client behavior, names, account context, and environment can influence the authentication path.

Pass-the-hash depends on a path where the captured material is usable. Relay depends on obtaining and forwarding an exchange to a destination that will accept it. If your observation shows that a different authentication mechanism was selected, an NTLM-specific hypothesis may not have been tested at all.

Signing and related integrity protections can break the relay path even when the client authenticates and the destination is reachable. At a conceptual level, signing binds protected traffic to the authenticated session so that an intermediary can’t freely forward or alter messages as though it were the authenticated party. The exact protection and requirement depend on the protocol and service configuration. For your decision, the important point is that a relay opportunity isn’t proven merely because you observed NTLM authentication. You must also establish that the intended destination’s configuration permits the relayed exchange to produce the requested operation.

Imagine that the workstation sends an eligible authentication exchange and your authorized relay test reaches the file server. The output reports that authentication was accepted, but the requested privileged operation fails. What did you prove?

You may have evidence that the exchange reached the service and that the service accepted the identity. You haven’t proved that the identity is an administrator, that code execution occurred, or that every service on the host is exposed to the same condition. Preserve the accepted identity, the destination service, and the failed operation as separate observations. Those boundaries keep a positive-looking line from becoming an unsupported compromise claim.

The opposite result also needs careful treatment. Suppose the connection reaches the file server, but the authentication attempt is rejected. That single result doesn’t establish why.

The material could be wrong for the claimed account. The account could be disabled or outside the expected scope. The target might not accept the relevant authentication method.

A signing or binding requirement might prevent the intended relay behavior. The request could have reached the wrong service, or your current network path could differ from the one you assumed. Failure narrows the investigation only when you know which prerequisite was actually exercised.

A disciplined workflow starts with a baseline, moves to one targeted test, and ends with independent validation. Record the access available before the test. Identify the exact authentication material and the identity you believe it represents.

Pass-the-hash uses suitable password-hash material in an authentication mechanism. Relay forwards a live authentication exchange under conditions that allow that forwarding to be accepted. A saved challenge-response record is not simply a substitute for the stored hash in a pass-the-hash test.

Confirm the authorized destination, reachable service, expected protocol, and desired result. Then perform one bounded test appropriate to the lab. Preserve the request or command you used, relevant output, and timing. Finally, check the resulting identity and capability through a direct observation rather than relying solely on the attacking tool’s success label.

That last check is essential because tools summarize complex exchanges. A message such as authentication successful can confirm one transition while saying nothing about the privileges granted afterward. A message reporting command execution may be stronger, but you still need to associate the output with the correct target and identity.

Conversely, a timeout doesn’t necessarily mean authentication was rejected. It could reflect routing, filtering, service availability, or a response the tool didn’t parse as expected. Read each line as evidence about one part of the chain, not as a verdict on the entire environment.

When the first attempt is ambiguous, find the prerequisite you assumed instead of observed. Suppose you believed the file server accepted NTLM, but you only confirmed that a file-sharing service was reachable. Those aren’t the same observation.

Use a lower-impact check permitted in the lab to isolate the service’s authentication behavior. If protocol selection is established but relay still fails, examine the relevant target protection and account context. If authentication succeeds but the desired operation fails, examine authorization. Change one part of the hypothesis, repeat the bounded test, and compare the new result with the original.

Don’t respond to uncertainty by cycling through unrelated tools. Changing the target, protocol, identity, syntax, and execution context at once may eventually produce a result, but it destroys your ability to explain why. It can also create unnecessary traffic outside the narrow question you were authorized to investigate. A failed test has value when your notes show its conditions. For example, documenting that the account authenticated to a service but lacked permission for a protected resource eliminates one explanation while preserving the possibility of different access within the same authorized scope.

The same reasoning applies to pass-the-hash troubleshooting. If suitable hash material works against one host but not another, don’t conclude that the second host is secure against every use of that identity. Compare the systems.

Is the account valid on both? Is it local on one and domain-based on the other? Is the expected protocol available?

Is the service reachable from your execution point? Does the account possess different rights on each destination? The hash material may remain the same while the target-specific conditions and resulting authorization change completely.

You should also resist the reverse mistake. Success against one service doesn’t prove success against the whole machine or the wider domain. It proves what your observed action proves.

If the test established access to a particular share as Lab Support, report that access. If you then independently established administrative execution, preserve the evidence for that separate transition. Each link should show the prerequisite, triggering action, resulting identity or context, and verified capability. That structure turns a collection of outputs into a defensible access path.

Good notes begin before success. Record the lab authorization and target identifiers, your starting host, current user context, the source and type of authentication material, and any uncertainty about its identity or scope. Preserve the destination address and service, the relevant configuration observations, the exact request or command, and the meaningful output. If an image is needed to support the result, make sure it captures the target context and evidence rather than only a generic success message. Keep the text needed to reproduce the sequence as well, since an image alone rarely explains every condition.

Your record should distinguish observation from interpretation. An observation might say that the destination accepted authentication for Lab Support and returned an access-denied response for a protected resource. Your interpretation might be that the account is valid for that authentication path but lacks the required resource permission. A broader claim that the account has no useful access would need more evidence. This separation helps another technically competent reader understand what happened, repeat the relevant steps, and challenge an assumption without reconstructing your entire session from memory.

Include failed attempts when they explain the corrected path. If your first relay test targeted a service that required a protection incompatible with the attempted relay, record the rejection and the evidence supporting that explanation. If you later test a different, explicitly authorized service after validating its conditions, record why that test answered a different question. Don’t rewrite the history as though the successful path was obvious from the start. The failed result demonstrates the condition that mattered and makes your final reasoning more reproducible.

Documentation also affects your next penetration-testing decision. If pass-the-hash authenticates an account but produces only limited access, the useful result may be new visibility rather than immediate control. If relay authenticates successfully but can’t perform the privileged operation you predicted, the result may still establish a configuration weakness and an account relationship worth reporting. If signing or another target protection prevents the path, that defensive condition is itself relevant evidence. In every case, choose the next action from the verified identity, privileges, reachable services, and remaining authorization, not from the excitement or disappointment of one tool message.

Keep the operating boundary explicit throughout. These techniques involve real authentication material and can affect other systems, so practice only with deliberately vulnerable targets, synthetic identities, or systems you own or have clear permission to test. Authorization should specify the environment and permitted activity. An assessment method is more than technical capability. It includes planning the test, limiting it to scope, analyzing the evidence, and reporting what the results mean without claiming more than they support.

That separation should influence how you practice. Automation can accelerate collection in a lab, but it shouldn’t replace your understanding of the conditions or your ability to preserve the underlying evidence. You should be able to explain what material was available, which protocol path was attempted, why the destination could or couldn’t accept it, what identity resulted, and what that identity was authorized to do. If you can’t answer those questions without quoting a tool’s final status line, the chain isn’t yet understood well enough to troubleshoot or report confidently.

Return to the original decision. You had authentication material and a reachable host, but that alone didn’t justify either technique. Pass-the-hash became a reasonable test only when the material, account, protocol path, and authorized destination aligned.

Relay became a reasonable test only when an eligible live exchange, reachable destination service, and compatible target protections aligned. In both cases, the result mattered only after you verified the resulting identity and access. That evidence now gives you a grounded next step, including the possibility that the right decision is to preserve the finding and investigate a different path rather than forcing this one further.

Pass-the-hash and relay require different material and protocol conditions, and both demand verification of the resulting identity and access. For your next review, ask yourself: Do you have the stored hash needed for pass-the-hash or a live eligible exchange and compatible destination for relay? The companion study reference includes a review checklist and an answer guide. This lesson was produced by Bare Metal Cyber dot com.

Separating Pass-the-Hash from Relay
Broadcast by