Pivoting, Relay, and Internal Discovery

You’ve gained authorized access to a workstation during a penetration test. From that workstation, you can reach a management subnet that wasn’t reachable from your original testing system. You also find evidence that the workstation communicates with several internal services.…

You’ve gained authorized access to a workstation during a penetration test. From that workstation, you can reach a management subnet that wasn’t reachable from your original testing system. You also find evidence that the workstation communicates with several internal services. What, exactly, have you gained? Is that enough to move to another host, or do you still need something else?

The careful answer is that you’ve gained a new network position and some possible leads. You haven’t automatically gained an identity that another host will accept. You haven’t proved that every reachable system is in scope. You also haven’t established that using any discovered credential or relaying any authentication is permitted. Reachability, authentication, authorization within the environment, and authorization under the engagement are separate questions.

Lateral movement means progressing from one accessed system or identity context to additional systems within an environment. The PenTest+ objective for this area includes pivoting, relay creation, discovery of services and network traffic, additional credential capture, credential dumping, and string searches. Those activities are related, but they’re not interchangeable stages of one automatic process.

Discovery tells you what may exist. Credential access may give you material associated with an identity. Pivoting may give you a path to a network that you couldn’t previously reach. Lateral movement occurs when an accessed position or identity is actually used to progress to another system. That separation matters because discovering a credential isn’t proof that it works, and proving that a service is reachable isn’t proof that you’re permitted to access it.

Pivoting uses an already-accessed system or network position to reach additional systems that were otherwise inaccessible from the tester’s original location. Imagine that your testing system can communicate with the employee network, while the accessed workstation has another permitted connection to a management subnet. If the engagement allows it, traffic can use that workstation as an intermediate point. The workstation changes where the traffic appears to originate and what network destinations it can reach.

That new vantage point doesn’t add identity privileges by itself. A management server may answer a connection from the workstation and still require an approved account. It may reject the credentials you currently possess. It may also be outside the authorized target list even though the route exists. A pivot provides reachability, not automatic authentication and certainly not permission to expand the engagement.

Relay creation uses an existing protocol interaction or trust relationship as a possible path toward another authorized target. Conceptually, rather than treating captured authentication material as though it were a reusable password, a relay attempts to pass an authentication interaction toward another service that may accept that context. Whether that is possible depends on the protocols and controls involved. Whether it is appropriate to test depends on the rules of engagement, the named targets, the allowed techniques, the timing, and the likely operational effect.

Pivoting and relaying therefore solve different parts of the problem. A pivot concerns network position and reachability. A relay concerns how an authentication interaction may be presented to another service. Neither one proves that the destination is authorized for testing, and neither should be treated as a general grant of access. Before either action, you need a prediction about what should happen, a bounded way to test it, and evidence that lets you compare the observed result with that prediction.

Start with the permitted objective, not the preferred tool. Which authorized objective are you trying to validate? Perhaps the engagement asks whether compromise of an ordinary workstation could expose administrative systems on a management subnet. That is more specific than a broad instruction to see how far you can go. It identifies the security question, but you still need to confirm the permitted source system, destination range, identity context, test period, and operational constraints.

Next, describe the expected state before taking stronger action. If the workstation is a usable pivot, you expect it to provide network reachability to a named, in-scope destination. That prediction says nothing yet about successful authentication. If a particular identity is expected to work at the destination, that is a second hypothesis and needs separate evidence. Keeping those predictions separate prevents a network failure from being mistaken for a credential failure, or a credential rejection from being mistaken for failed routing.

Then choose the least-disruptive action that can establish the needed fact. If passive records or an already-approved connectivity check can confirm that the workstation reaches the destination, there’s no reason to begin with a more invasive action. If stronger validation is necessary, it still has to remain within the specified target, time, identity, and operational limits. A technically available action isn’t automatically the right engagement decision.

Consider the workstation and management subnet again. The workstation can route traffic toward a named management server. Your first supported conclusion is narrow: a network path appears to exist from the accessed workstation to that destination. You can’t yet report that the management server is compromised, that the current credentials are valid there, or that the whole management subnet is exposed. Those conclusions require different evidence.

Suppose the approved validation confirms that the named service answers through the workstation. You’ve now corroborated reachability. If an authorized identity is rejected, that result doesn’t invalidate the pivot.

It tells you that the network path and the identity outcome differ. The pivot may be functioning correctly while the destination refuses authentication. Treating those results as one event would hide the very control boundary the test is meant to examine.

Discovery helps you decide which hypothesis is worth testing. Service discovery identifies listening or responsive services associated with authorized targets. Network-traffic discovery examines communications visible from the reached position.

Those observations may reveal which systems communicate, what services appear to be involved, and where an authentication exchange may occur. Packet analysis can organize and filter captured traffic, but a visible packet or protocol label still needs interpretation. It isn’t proof of compromise on its own.

The same restraint applies to credential discovery. Additional credentials may appear in authorized data sources available from the accessed context. A credential dump is an attempt to obtain credential material from a system or process, while a string search looks through permitted data for readable sequences that may include usernames, tokens, passwords, connection details, or other useful clues. These activities can expose highly sensitive material, so their use should be explicit in the engagement boundaries rather than assumed from general system access.

A string that resembles a password is an observation, not a validated credential. It could be old, incomplete, unrelated to the intended account, or unusable at the destination. Credential material recovered through a dump also doesn’t prove that it is valid for another system. It may help form a hypothesis about identity reuse or privilege relationships, but lateral movement hasn’t occurred merely because the material was found.

Network traffic can be similarly ambiguous. You may observe that the workstation communicates with a management server, but that doesn’t prove the workstation can initiate every kind of connection to it. You may see an authentication-related exchange without knowing whether it can be relayed or whether the destination would accept it. Data freshness matters too. An old capture, stale service record, or outdated host information can produce a convincing but incorrect path.

When evidence conflicts, troubleshoot the assumptions before escalating. Confirm that you’re examining the intended source and destination. Check whether the target remains in scope and whether the relevant test period is still open.

Revisit the expected network path, the identity context, prerequisites, and the quality of the observation. A failed tool output might reflect configuration or data quality rather than a security control. A successful connection might prove only that a service responded, not that access was granted.

Suppose the route through the workstation is confirmed, but an attempted connection to the authorized management server doesn’t succeed. One tempting conclusion is that the pivot failed. Another is that the credentials are invalid.

Neither follows without more evidence. The failure could involve the destination selection, the service, the network path, the identity, or a dependency that wasn’t satisfied. The next validation should isolate one of those possibilities without broadening the test.

A lower-impact reachability check may distinguish the path from the authentication attempt. If reachability is confirmed again, you can focus on the service and identity context. If the service responds but rejects the approved identity, record that authentication was unsuccessful under that specific context. Don’t silently substitute another account, scan unrelated systems, or capture additional credentials merely because those actions might produce an answer. Escalation or clarification is the correct next step when the authority is unclear.

Now consider a different outcome. The service accepts an approved identity through the pivot. That validates more than reachability, but the wording still matters.

You’ve demonstrated that this source position, destination service, and identity context can form a usable path under the test conditions. You haven’t established that every user, protocol, or destination on the subnet behaves the same way. A defensible finding describes what the evidence actually supports and preserves any unresolved uncertainty.

A path log makes that discipline practical. For each attempted or validated hop, record the source, the destination, the credential or execution context, and the authorization that covers the action. The source may be your original testing system or an already-reached host.

The destination should be the specific in-scope system or service being tested. The context identifies which approved identity or system position was used. The authorization entry ties the hop to the applicable target, technique, timing, and operational restrictions.

Add the predicted result, the action taken, the observed evidence, and the conclusion that evidence supports. If the hop fails, preserve that result rather than erasing it from the path. Failed attempts can expose a mistaken assumption, an effective control, or a test condition that remains unresolved. The log should also distinguish a planned hop from an attempted one and an attempted hop from a validated one.

For example, the first entry might show the testing system as the source and the employee workstation as the destination, with the authorized access context that established the initial position. A second entry might show the employee workstation as the source and the named management server as the destination. Its context would identify the approved identity used, while its evidence would separately state whether network reachability and authentication were confirmed.

If the second hop used a relay rather than directly supplied credentials, the log should preserve that distinction. The relevant context is the authentication interaction being tested, not an invented claim that a password was known. If the result shows only that the destination responded, record a response. If it shows accepted authentication under the approved conditions, record that narrower conclusion. Accurate context helps reviewers understand what control failed or held.

This path record also supports cleanup and reporting. It shows which systems were involved, which identities or contexts were exercised, and where sensitive credential material may have been handled. It lets the tester explain the chain without relying on memory and helps the organization identify the most important corrective points. An attack path can involve several connected privileges or behaviors, so remediation may focus on breaking a high-value link rather than treating every host as an isolated event.

Imagine you must choose between two next actions. The first is to conduct broad credential capture across every host visible from the pivot. The second is to validate reachability to one named server using an already-approved context, then request clarification if stronger testing is needed. Both might produce information, but only the second follows the evidence and the stated boundary. The first silently turns network visibility into permission for a wider and more intrusive test.

Additional information could change that decision. If the rules of engagement explicitly name credential dumping, identify the target hosts, set an approved window, and address operational impact, a carefully bounded credential-access test may become appropriate. If those details aren’t present, technical feasibility doesn’t fill the gap. Authorization is part of technical correctness in professional penetration testing, not an administrative concern added after the work.

The same reasoning applies when a relay opportunity appears in network traffic. Before testing it, identify the in-scope source interaction, the intended destination, the expected authentication outcome, and the evidence needed to establish the security issue. Consider whether a lower-impact method can demonstrate the relevant trust problem. If direct validation is required, constrain it to the approved identity, service, target, and time. Then compare the result to the prediction instead of treating any tool message as final truth.

Reporting should preserve the difference among discovery, credential access, and actual movement. You might report that the accessed workstation could reach a management service, that approved testing identified credential material, and that a specific identity was or wasn’t accepted by a named destination. Those are separate evidence statements. Combining them into a claim that the entire environment was traversable would go beyond the observations unless every part of that broader claim had been validated.

This distinction affects remediation. A reachable management network may indicate a segmentation issue. Accepted authentication may point toward an identity or privilege problem. Sensitive strings or credential material found in an accessed context may indicate exposure that needs correction even if no lateral movement succeeds. Separating the evidence helps the organization address the control that actually matters rather than receiving one vague statement that internal access was possible.

Return to the original workstation. You know it can reach a management subnet, and you know it communicates with internal services. That gives you a vantage point and a set of hypotheses.

Your next decision comes from the authorized objective, the precise target, the identity or protocol context, and the evidence still missing. Record each hop before moving on. That keeps a possible path from being mistaken for a proven one and turns lateral movement testing into a controlled chain of decisions rather than an unbounded search for access.

Pivoting, credential discovery, and lateral movement answer different questions. A pivot may change what a reached system can communicate with, while credential evidence may suggest an identity worth validating. Neither one proves that another host can be accessed.

That conclusion requires a bounded test, corroborating evidence, and a destination that is explicitly authorized. Keep each hop tied to its source, destination, identity or protocol context, observed result, and engagement authority. When those details stay separate, you can distinguish a possible route from a validated attack path, explain exactly what the test established, and give the organization evidence it can use for focused remediation.

Before moving on, ask yourself: Is that enough to move to another host, or do you still need something else? Use the companion episode reference for exact terminology, named tools or standards, source-derived evidence checkpoints, and review material for Pivoting, Relays, Service Discovery, and Credential Capture. This lesson was produced by Bare Metal Cyber dot com.

Pivoting, Relay, and Internal Discovery

headphones Listen Anywhere

More Options »
Broadcast by