Using DHCP Snooping, DAI, and Port Security

This episode focuses on Using DHCP Snooping, DAI, and Port Security. It develops three connected ideas: DHCP snooping establishes trust and binding evidence; DAI evaluates ARP mappings; and port security controls source MAC use on an edge port. The lesson uses prediction, observ…

Today's lesson is Using DHCP Snooping, DAI, and Port Security. Two engineers are investigating the same access switch. A legitimate computer can’t communicate normally, and both engineers suspect a Layer Two security feature. One wants to trust the computer’s switch port.

The other wants to increase the number of secure MAC addresses allowed on that port. Those changes affect different controls, so choosing between them by instinct could hide the real problem or weaken the intended protection. Before changing anything, they need evidence. Is the switch rejecting DHCP traffic, is an ARP mapping failing inspection, or has port security detected a source MAC address that the port isn’t allowed to use?

As the scenario develops, listen for three decisions: DHCP snooping establishes trust and binding evidence; DAI evaluates ARP mappings; and port security controls source MAC use on an edge port. The objective is not to hold every displayed value in memory. It is to understand why each value matters and what conclusion it supports. Start with the relationship that controls the result.

Those questions separate three features that appear together in the CCNA objective for configuring and verifying Layer Two security. DHCP snooping, Dynamic ARP Inspection, and port security all operate at the switching edge, but they don’t establish the same kind of trust. DHCP snooping controls which paths may carry trusted DHCP server behavior and learns binding information from legitimate DHCP activity. Dynamic ARP Inspection, often shortened to D AI, checks ARP information against trusted binding data on inspected paths.

Port security controls which source MAC addresses may use a particular switch port and defines what happens when that policy is violated. Begin with DHCP snooping. Imagine an access VLAN containing employee computers. The legitimate DHCP server is reached through an uplink, while individual computers connect through access ports.

The switch needs to distinguish the path toward the legitimate server from paths where ordinary clients are connected. The uplink carrying valid server messages can be treated as trusted for DHCP snooping. The client-facing access ports remain untrusted. Here, trusted and untrusted describe how the switch handles DHCP messages arriving on those paths.

They aren’t broad declarations that everything received through an uplink is safe or that every client is malicious. That distinction lets the switch reject invalid DHCP behavior arriving where server behavior isn’t expected. It also lets the switch build a snooping binding database from legitimate DHCP activity. A binding associates information learned through that exchange with the client’s location.

The important mental model is evidence created from observed, accepted activity. Enabling the feature is only the configuration input. The resulting binding state is what another security decision may depend on. If the expected host binding isn’t present, simply seeing that DHCP snooping is enabled is not enough to establish the switch learned what you expected.

This dependency matters. If a legitimate host has no usable binding, D AI may reject that host’s legitimate ARP information even though the host itself did nothing wrong. The inspection feature is behaving according to the evidence it has. The missing or incorrect evidence is the problem.

Conversely, a healthy snooping database doesn’t mean ARP inspection must be active everywhere it’s needed. One feature can supply information that another feature uses, but supplying the information and enforcing the ARP decision are separate actions. Port security asks another question entirely. Instead of asking whether a DHCP message arrived through an appropriate path or whether an ARP mapping agrees with trusted data, it asks which source MAC addresses are permitted to use a switch port.

A port can have a secure MAC policy and a limit on the addresses allowed there. Secure addresses may be established according to the configured learning method, including static or sticky learning contexts. If traffic arrives with a source MAC address that violates the port’s policy, the configured violation behavior determines the consequence. Suppose a desk port is intended for one computer and the port has reached its allowed secure MAC limit.

Someone replaces the computer, and the replacement sends frames with a different source MAC address. DHCP might be available through the correct trusted path, and the replacement might request perfectly legitimate addressing information. Its ARP mapping might also be internally consistent. Port security can still reject its use of the port because the new source MAC address isn’t allowed by the port’s current secure state.

Trusting a DHCP path wouldn’t correct that mismatch. When you are ready to work the example, open the companion reference for source-derived detail related to binding tables, trust-port diagram, and violation output. Do not treat it as a second narration script. Use it to perform the calculation, configuration, or interpretation while applying the reasoning developed here.

Now follow that relationship into the operating state. Now apply all three controls to the employee VLAN. The uplink toward the legitimate DHCP service is expected to be trusted for DHCP snooping. User-facing access ports are expected to be untrusted.

DHCP snooping observes accepted client activity and builds bindings. D AI inspects ARP information on the protected client paths and compares it with trustworthy binding data. Port security controls the source MAC addresses allowed on each access port. Before checking the switch, you should be able to predict those roles and the state each feature should create.

A legitimate computer connects to one of those access ports. First, its DHCP exchange must travel through the expected client and server paths. If the server-facing path is incorrectly left untrusted, legitimate server traffic can be filtered as invalid. The computer may fail to receive the expected addressing information, and the switch may also fail to build the expected binding.

That one trust error can produce more than one symptom because the missing DHCP result also removes evidence that D AI may need later. Notice what port security contributes in that situation. It might show that the computer’s source MAC address is permitted and that no port-security violation has occurred. That evidence doesn’t repair the DHCP path, but it helps eliminate one competing explanation.

Increasing the secure MAC limit would be an unsupported change because the operational state hasn’t shown a MAC limit or secure-address violation. The right investigation stays with the trusted and untrusted DHCP roles and the binding state they’re expected to produce. After the trust role is corrected, don’t treat acceptance of a configuration change as proof that the network is healthy. Confirm that legitimate DHCP activity now succeeds through the intended path.

Then check whether the expected binding appears. That sequence connects configuration to operational state. If the host works but no expected binding is present, the outcome still doesn’t match the prediction. If the binding appears and the host’s ARP information passes inspection, the observed results support the causal chain you expected.

Now consider a different failure on the same VLAN. DHCP succeeds, the host has the expected binding, and D AI accepts its valid mapping. Frames from the host still trigger a port-security violation because the source MAC address doesn’t match the port’s permitted secure state or exceeds the applicable limit. In this case, changing the DHCP trust role would be irrelevant and potentially harmful.

The binding and inspection evidence already show that those mechanisms are doing their jobs. The mismatch is between the source MAC address and the state enforced on the access port. Troubleshooting becomes much safer when you write the expected behavior in your own mind before making a change. For the protected VLAN, you expect server behavior through the trusted DHCP path, client behavior through untrusted access ports, bindings created from legitimate DHCP activity, ARP mappings validated on inspected paths, and source MAC addresses constrained at the edge.

Those predictions tell you which verification evidence matters. Without them, a screen full of status information can encourage random changes rather than diagnosis. Use the evidence to distinguish the competing explanations. Start with the symptom, but don’t stop there.

If the computer can’t obtain addressing information, inspect the DHCP trust relationship and the binding result. Ask whether legitimate server traffic is arriving on a path the switch treats appropriately. If the computer has addressing information but local communication fails when it sends ARP information, examine the D AI result and the binding evidence used for that decision. If the switch reports a port-security violation, compare the observed source MAC address and port state with the intended secure policy.

Configuration text alone can’t answer all of those questions. It tells you what someone intended to request from the switch. Operational state tells you what the switch learned and what it’s enforcing now. For DHCP snooping, that includes the trusted or untrusted role of the relevant path and the resulting bindings.

For D AI, it includes whether inspected mappings are accepted or dropped against the available evidence. For port security, it includes secure MAC state, applicable limits, and whether a violation has occurred. This is especially important when a security control blocks legitimate traffic. It’s tempting to conclude that the protection is broken and disable it.

A better first question is whether the protection received the correct trust assignment or state evidence. If D AI rejects a valid host because its expected binding is absent, turning off inspection removes the symptom but doesn’t explain the missing evidence. If port security rejects a replacement computer because the secure port state still reflects a different source address, trusting the port for DHCP doesn’t address the violation. Change one supported cause, then repeat the same verification.

If you correct a DHCP trust role, verify legitimate DHCP behavior and the resulting binding before evaluating D AI again. If you correct the secure MAC state or limit to match an approved port requirement, verify the port-security result rather than merely waiting for the computer to appear functional. Repeating the check matters because a disappearing symptom isn’t always proof that the intended protection remains active and correctly enforced. Imagine that both engineers from the beginning now collect their evidence.

They find no port-security violation, and the host’s source MAC address is already represented in the permitted state. Their proposed increase to the address limit has no factual support. They also find that the host’s expected binding is absent. The legitimate DHCP server is reachable through a path treated as untrusted, so valid server behavior isn’t producing the expected result.

The trust-role correction fits the observed mismatch much better than a port-security change. After correcting that path, the engineers observe accepted DHCP activity and the expected binding. D AI can now compare the host’s ARP information with trustworthy data, and the valid mapping passes. Port security continues to enforce the original source MAC policy without alteration.

The repair succeeds because it restores the missing dependency rather than weakening every Layer Two protection that happened to be near the symptom. Finish by choosing the smallest correction the evidence supports. A different set of evidence could justify the other engineer’s area of concern. Suppose the DHCP exchange succeeds, the binding is present, and valid ARP information passes inspection.

The switch then reports that the access port has encountered a source MAC address outside its allowed secure state. That result directs attention to port security. You’d compare the intended number and identity of source MAC addresses with what the port currently permits, then apply only a correction that matches the approved use of that port. That’s why enabling any one of these features doesn’t automatically supply the protection of the other two.

Port security can constrain source MAC addresses without creating the DHCP binding evidence D AI commonly relies on. DHCP snooping can build bindings without itself applying the port’s secure MAC limit. D AI can reject an invalid mapping, but that doesn’t define which source addresses a port may learn or what violation behavior port security should apply. They cooperate, but they remain distinct mechanisms.

For configuration and verification, keep a three-part chain in mind. First, state the requirement in operational terms. Identify which path should be trusted for legitimate DHCP server behavior, where ARP inspection should use binding evidence, and which source MAC addresses the edge port should permit. Second, predict the state.

Name the bindings, inspection results, and secure port conditions you expect. Third, compare that prediction with actual evidence and correct the smallest mismatch that explains the observation. This approach also keeps your work aligned with the published CCNA objective. The objective calls for configuring and verifying DHCP snooping, Dynamic ARP Inspection, and port security.

Verification means more than recognizing their names or seeing configuration statements. You need to connect each feature to the Layer Two relationship it protects, predict the state it should produce, and interpret the observed operating state when a legitimate host is blocked. When a failure appears, resist broad labels such as Layer Two security problem. That label is too large to select a correction.

Determine whether the contradiction concerns a DHCP path, a missing or untrustworthy binding, an ARP mapping under inspection, or a source MAC policy on the access port. Once you can name the failed relationship, the relevant evidence and the justified change become much clearer. Bring the evidence together. DHCP snooping, Dynamic ARP Inspection, and port security work together without becoming interchangeable.

Snooping establishes trusted and untrusted DHCP paths and builds binding evidence. D AI can use trustworthy bindings to judge ARP mappings on inspected paths. Port security separately controls which source MAC addresses may use an edge port and what happens when that policy is violated. When legitimate traffic is blocked, don’t weaken all three controls or choose a fix from the symptom alone.

Predict the required state, inspect the relevant trust role, binding, inspection result, or secure port state, and change only the cause supported by that evidence. Knowledge check: Which control should you investigate when a legitimate MAC address is blocked on one access port? Stop for retrieval.

Give the answer without notes, then check whether you included both the technical relationship and the evidence needed to verify it. Thanks for listening. This lesson was produced by Bare Metal Cyber dot com.

Using DHCP Snooping, DAI, and Port Security
Broadcast by