Extending VLANs Across Switches

This CCNA lesson covers Extending VLANs Across Switches. The discussion centers on three points: local VLAN membership is only one part of the path; the interswitch link must carry the VLAN; and traffic between VLANs requires Layer Three forwarding. It connects the mental model…

Today's lesson is Extending VLANs Across Switches. The goal is to build a model you can recall without looking at a screen. Start with this principle: Local VLAN membership is only one part of the path. Use the episode reference later for the exact characters, rows, commands, or values needed in a lab.

As the scenario develops, listen for three decisions: local VLAN membership is only one part of the path; the interswitch link must carry the VLAN; and traffic between VLANs requires Layer Three forwarding. 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.

Two engineers are investigating the same network failure. One says the switches need matching virtual local area networks, or VLANs. The other says the connection between the switches is the problem. Both explanations sound reasonable, and either engineer could make configuration changes.

What evidence should decide which change, if any, is justified? To answer that well, you need to separate three conditions that are easy to blur together. A VLAN must exist where it’s needed. The correct access ports must belong to it.

And the path between switches must carry its traffic. Matching VLAN numbers are part of the design, but they aren’t the entire path. Picture two access switches in different parts of an office. A computer named Sales One connects to the first switch, and Sales Two connects to the second.

Both computers are intended to belong to VLAN ten. VLAN membership is configured locally on each switch. The first switch doesn’t automatically inherit the second switch’s VLAN information just because the devices are connected. Each switch needs the appropriate local state so that its access port places the attached computer into VLAN ten.

The interswitch connection also has to carry VLAN ten. If all of those conditions are met, the two computers can participate in the same Layer 2 forwarding domain even though they aren’t attached to the same physical switch. Layer 2 forwarding is based on Ethernet behavior, including media access control addresses, commonly called MAC addresses. The switches learn where devices can be reached and forward Ethernet frames toward the appropriate ports.

A broadcast sent within VLAN ten can also cross the interswitch path and reach VLAN ten ports on the other switch when that path carries the VLAN. That’s why the VLAN is described as one broadcast domain spanning multiple switches. The physical switches are separate boxes, but the logical Layer 2 boundary follows the VLAN and the path permitted to carry it. This distinction prevents a common mistake.

Creating VLAN ten on both switches doesn’t, by itself, make VLAN ten continuous between them. Imagine that Sales One sends traffic intended for Sales Two. The first switch accepts the frame on an access port assigned to VLAN ten. It then needs a valid Layer 2 path toward the second switch.

If the interswitch connection doesn’t carry VLAN ten, the forwarding process can’t simply reinterpret that frame as belonging to another available VLAN. The matching number on the far switch doesn’t bridge the missing section. From an end-to-end perspective, VLAN ten has been configured at both edges but interrupted in the middle. The reverse is also important.

An interswitch connection capable of carrying VLAN ten doesn’t repair incorrect access-port membership. If Sales Two’s port belongs to VLAN twenty, traffic arriving for VLAN ten won’t be delivered through that port as though the two VLANs were interchangeable. You therefore need to reason about local membership and interswitch carriage as related but distinct conditions. When same-VLAN communication succeeds, the switches have preserved the traffic’s VLAN association across the Layer 2 path.

The computers don’t need a routing decision merely because two switches are involved. The number of physical switches doesn’t determine whether routing is required. The logical VLAN boundary does. The companion reference preserves source-derived technical detail related to multi-switch VLAN path diagram and verification output.

Use it at a screen or in a lab rather than trying to memorize the exact display by ear. Read one requirement at a time, predict the expected state, and compare that prediction with the artifact in front of you. Now follow that relationship into the operating state. Now add a third computer named Finance One.

It connects to the second switch and belongs to VLAN twenty. Sales One in VLAN ten wants to communicate with Finance One in VLAN twenty. This is no longer the same kind of forwarding problem. VLAN ten and VLAN twenty are distinct Layer 2 broadcast domains.

A switch operating only at Layer 2 doesn’t take an Ethernet frame from one VLAN and silently forward it into another. Doing so would erase the separation that the VLAN design created. Communication between those VLANs requires a Layer 3 forwarding function, which makes a routing decision between their networks. That gives you a useful decision rule.

Ask whether the source and destination are in the same VLAN. If they are, investigate the Layer 2 path for that VLAN, including local port membership and whether the interswitch path carries it. If they’re in different VLANs, a complete explanation must include a Layer 3 forwarding function and a reachable gateway path. The fact that the devices happen to connect to one switch or two switches doesn’t change that rule.

Two hosts on different switches can communicate entirely through Layer 2 when they share a VLAN. Two hosts on the same physical switch still require Layer 3 forwarding when they belong to different VLANs. This also explains why a successful same-VLAN test proves less than people sometimes think. Suppose Sales One reaches Sales Two in VLAN ten.

You’ve shown that VLAN ten membership and carriage are working for the tested path. You haven’t proved that VLAN twenty is carried correctly. You also haven’t proved that routing between VLAN ten and VLAN twenty works. A single successful result validates only the conditions that result actually exercised.

Good verification connects each test to a specific forwarding claim instead of treating general connectivity as one indivisible state. Verification works best when you divide it into layers. First confirm VLAN membership at the edge. Is the computer connected to the expected access port, and is that port operationally associated with the intended VLAN?

The switch’s VLAN status information, including the information summarized by the VLAN summary display on Cisco switches, can help confirm local VLAN and port membership. This check matters on each relevant switch because VLAN membership is local state. Seeing VLAN ten on the first switch tells you nothing conclusive about VLAN ten’s state on the second. Next confirm interswitch carriage.

The key question is whether the connection between the switches actually carries the VLAN needed by the failed flow. Don’t infer that answer only from the fact that some traffic crosses the link. The link might carry one VLAN while failing to carry another required VLAN. If VLAN twenty works across the link but VLAN ten doesn’t, the physical connection clearly isn’t completely down.

The difference directs your attention toward the VLAN-specific operational state of that interswitch path rather than toward a broad physical-link failure. Use the evidence to distinguish the competing explanations. Then confirm the Layer 3 gateway path when the source and destination are in different VLANs. A correct access-port assignment doesn’t create inter-VLAN forwarding.

A working interswitch path doesn’t create it either. Those conditions can deliver the traffic to the appropriate Layer 2 boundary, but a Layer 3 function must still forward between the networks represented by the VLANs. Verification therefore needs to distinguish reaching the local VLAN from crossing into the destination VLAN. Otherwise, you may keep changing Layer 2 settings while the actual missing condition is Layer 3 forwarding.

Return to the original failure. Sales One and Sales Two are both supposed to be in VLAN ten, but they can’t communicate. An engineer checks the first switch and confirms that VLAN ten exists and Sales One’s port belongs to it. That’s useful evidence, but it isn’t enough to declare the configuration correct.

The engineer then checks the second switch. VLAN ten exists there too, and Sales Two’s access port has the expected membership. The tempting conclusion is that both edge configurations match, so the problem must be somewhere else. Yet the interswitch path still hasn’t been verified.

Evidence from the running network shows that the connection between the switches isn’t carrying VLAN ten. That single mismatch explains the symptom without contradicting either access-port check. Sales One’s frame enters VLAN ten correctly, but the Layer 2 path for VLAN ten doesn’t continue to the second switch. The appropriate correction is the one supported by that evidence.

Correct the interswitch carriage for VLAN ten, then repeat the same communication test and the same operational checks. If the test succeeds and the relevant state now matches the prediction, you’ve demonstrated why the correction worked rather than merely observing that the symptom disappeared. Changing several unrelated settings at once would weaken that conclusion. If you changed both access ports, recreated the VLAN, altered the gateway, and modified the interswitch connection, a successful test wouldn’t tell you which change repaired the path.

It could also introduce a second fault. Isolating the smallest state difference is both safer and more informative. Start with the expected behavior, inspect the evidence that represents it, correct one supported cause, and verify again. That approach turns troubleshooting into a test of the forwarding model instead of a sequence of guesses.

Suppose the local membership checks are correct for all three computers, and each required VLAN is carried across the relevant interswitch path. Same-VLAN communication works in both VLANs, but communication between VLAN ten and VLAN twenty still fails. The remaining distinction is the boundary between switching and routing. Layer 2 forwarding has been demonstrated within each VLAN.

It hasn’t demonstrated a Layer 3 path between them. You’d now focus verification on the gateway function and the path through that Layer 3 boundary. Creating more copies of VLAN ten or VLAN twenty wouldn’t address the missing type of forwarding. Finish by choosing the smallest correction the evidence supports.

You can also reason in the opposite direction. If inter-VLAN communication works, then some Layer 3 forwarding function is present in the successful path. That doesn’t mean every same-VLAN path across every switch is healthy. A different access port may have incorrect membership, or a different interswitch path may not carry the needed VLAN.

Again, evidence has a scope. The strongest conclusion is the one that follows directly from the tested source, destination, VLANs, and forwarding boundaries. Broad claims such as the VLANs are working or the network is routed hide the exact state you need to understand. For configuration and verification, translate the requirement into conditions before thinking about individual commands.

A requirement that Sales users on both switches share VLAN ten implies local VLAN state on both switches, access ports assigned to VLAN ten, and an interswitch path that carries VLAN ten. A requirement that Sales in VLAN ten communicate with Finance in VLAN twenty adds a Layer 3 forwarding requirement. The first requirement extends one Layer 2 broadcast domain across the switches. The second connects two separate broadcast domains without merging them into one VLAN.

That last distinction matters because routing doesn’t make VLAN ten and VLAN twenty the same broadcast domain. It provides controlled forwarding between their Layer 3 networks while the VLAN boundaries remain distinct. Within VLAN ten, switches handle the Layer 2 path. Within VLAN twenty, switches do the same for that VLAN.

At the boundary, the gateway handles the Layer 3 decision. When you trace a flow, identify which part is occurring within a VLAN and the exact point at which traffic must cross from one VLAN to another. That tells you what kind of forwarding evidence to seek. The CCNA objective expects configuration and verification of normal-range VLANs spanning multiple switches, along with inter-VLAN connectivity.

Those verbs matter. It isn’t enough to recognize that two hosts have matching VLAN numbers. You need to connect the requirement to operational state and determine whether the forwarding path supports it. At the same time, this episode’s boundary is the decision model rather than the detailed mechanics of every interswitch or gateway implementation.

The durable skill is knowing what each part of the network must accomplish and how to prove which part failed. So when two proposed fixes compete, don’t choose the one with the most familiar command or the broadest change. Classify the traffic first. For a same-VLAN flow across switches, inspect local membership and carriage of that VLAN across the Layer 2 path.

For a different-VLAN flow, verify those Layer 2 conditions and then verify the Layer 3 gateway path. If the evidence shows correct edge membership but interrupted interswitch carriage, fix the interruption. If Layer 2 works within both VLANs but traffic can’t cross between them, investigate the routing boundary. The observed state, not the matching labels alone, decides which explanation survives.

Bring the evidence together. A VLAN can span more than one physical switch because its logical boundary follows the Layer 2 path that carries it. That requires correct local membership and uninterrupted carriage between the switches. Matching VLAN numbers at the edges aren’t enough when the middle of the path is missing.

Communication between different VLANs is a different case. It must cross a Layer 3 forwarding boundary rather than remain inside one broadcast domain. When connectivity fails, verify those responsibilities separately. Confirm edge membership, confirm interswitch carriage, and confirm the gateway path when routing is required.

That evidence lets you correct the actual broken boundary instead of making several speculative changes and hoping one of them works. Knowledge check: Why can matching VLAN numbers on two switches still fail to provide end-to-end Layer Two connectivity? Pause here and answer in your own words.

Name the relevant relationship, the evidence you would inspect, and the conclusion that evidence would support. Thanks for listening. This lesson was produced by Bare Metal Cyber dot com.

Extending VLANs Across Switches

headphones Listen Anywhere

More Options »
Broadcast by