Advertisement

VLAN network not getting a DHCP address: find the break

When a VLAN client never gets a lease, the break is usually in one of five places: the client, VLAN tagging, the trunk or relay, the DHCP scope, or a filter blocking replies. This article helps you prove where DHCP stops before you change anything.

Networking 8 min read
A managed office Ethernet switch sits in a wiring closet with network cables plugged in.

When a VLAN office network won’t get a DHCP lease, the break is usually in one of five places: the client, the access port or VLAN tagging, the trunk or gateway relay, the DHCP scope, or a filter such as ACLs, firewall rules, or DHCP snooping. The quickest fix is to find where the DHCP Discover stops, then change only that piece. A broadcast from one VLAN does not reach another VLAN on its own, and DHCP depends on those broadcasts.

Symptom

The usual symptom is a device that sits on “obtaining IP address,” falls back to an automatic or private address, or never finishes renew after a lease expires. The key detail is not just that DHCP failed. It is whether the failure hits one device, one switch port, or the whole VLAN.

If one host fails while others in the same VLAN work, the problem is usually local to that device, its port, or its NIC settings. If every host in the VLAN fails, the fault is more likely in the VLAN path, the relay, the DHCP scope, or an upstream filter.

A static IP test helps only with a narrow question. It proves the client can talk to the gateway and local network at layer 3. It does not prove DHCP is working, because request and reply traffic still has to cross the right VLAN and relay path.

Before you start

  • Have access to the client, the switch, and the DHCP server or relay device.
  • Know which VLAN the client is supposed to be in, and which interface on the switch or router carries that VLAN.
  • If the site uses router-on-a-stick or a Layer 3 switch, know where the DHCP relay or helper address is configured.
  • If the site uses managed Wi-Fi or a hypervisor, confirm whether the SSID or virtual port is mapped to the same VLAN you expect on the wired side.
  • If you are about to change relay, firewall, or snooping settings, do one change at a time so you can roll it back if needed.

The cleanest way to troubleshoot this is to answer one question at a time. Did the client send a DHCP Discover? Did it leave the VLAN? Did anything block the reply on the way back? A DHCP client asks for an address by broadcast, and if that request never reaches the DHCP server, no lease is handed out.

1. Prove whether the DHCP Discover leaves the VLAN at all

Start on the client, or as close to it as you can get. Renew the lease and watch for a DHCP Discover leaving the NIC. If you can packet-capture on the client or the access switch, that is the fastest way to separate a client problem from a network problem.

  1. Renew the lease and watch for a DHCP Discover leaving the NIC.
  2. Capture on the client or the access switch if you can.
  3. Use what you see to decide whether the client or the network is at fault.
  • If you see Discover frames leave the client side, the client is at least trying DHCP.
  • If you do not see Discover frames leave, check the NIC, OS DHCP setting, local firewall, port security, or the wrong VLAN assignment before you touch the server.
  • If the Discover leaves but nothing comes back, the break is upstream: relay, scope, firewall, ACL, or snooping.

This is also where a static IP test has limits. If a static address works, you know the port and gateway path can pass normal traffic. You still do not know whether broadcasts are reaching the right VLAN or whether the relay can forward DHCP requests.

An Ethernet cable is plugged into a port on an office Ethernet switch.

2. Verify VLAN membership, access port mode, and trunk allowed VLANs

Next, confirm the port and trunk are actually carrying the VLAN you think they are. A live link can still be in the wrong access VLAN, or the uplink trunk can simply omit the VLAN.

  1. Check the access port and confirm it is in the intended access VLAN, not the default or native VLAN by mistake.
  2. Check the uplink and confirm the VLAN is allowed across the trunk and tagged the way the downstream device expects.
  3. If there is a Layer 3 switch or router-on-a-stick path, verify the VLAN interface or subinterface for that subnet is present and active.

If only one host fails, this step is still worth checking. The odds move toward a bad patch cable, a wrong port profile, a bad NIC, or port security. If every host in the VLAN fails, a missing VLAN on the trunk or a wrong access VLAN is one of the first things I would rule out. In practice, the edge port can look correct while the uplink blocks that VLAN one hop away.

An office Ethernet switch is mounted in a server rack with uplink cables connected.

3. Check DHCP relay or helper settings on the gateway

If the VLAN is correct, check the place where DHCP is supposed to leave the subnet. On router-on-a-stick designs and Layer 3 switches, that is usually the SVI or subinterface with the helper address or relay setting.

  1. Confirm the helper points to the real DHCP server for that VLAN.
  2. Confirm it is on the interface for the correct subnet, not on a different VLAN interface by accident.
  3. Confirm the relay path is not tied to the wrong VRF or wrong server address.

A scope on the DHCP server does not help if the relay is missing or pointing at the wrong place. If you can see client broadcasts on the VLAN but never see the server answer, relay is one of the first things to check. The DHCP server can only hand out an address when it receives the request.

4. Check scope exhaustion, exclusions, ACLs, firewall rules, and DHCP snooping

Once the VLAN and relay look right, check the server and the filters on the path.

  1. Confirm the scope for that subnet exists and is active.
  2. Check whether the scope is exhausted.
  3. Review exclusions and reservations that may have consumed the pool.
  4. Make sure any ACLs or firewall rules on the routed path allow DHCP traffic.
  5. Check DHCP snooping or similar edge protection on the switch, especially if offers seem to arrive at the switch but never reach the client.

Scope exhaustion is easy to miss because the network may still look healthy for traffic that already has an address. ACLs and firewall rules can be just as deceptive. Ping or web traffic may work while DHCP fails, because DHCP uses broadcast during discovery and uses UDP ports that are easy to overlook. Snooping often breaks the return path, which means the reply can reach the switch and still be dropped before it reaches the client.

A common clue is this. If the client’s Discover leaves, but no Offer ever reaches the client, the likely causes are a bad helper, a full scope, or filtering on the return path. If the Offer reaches the switch and still dies there, focus on snooping or another switch-side filter.

5. Confirm the fix

After the change, renew the client lease and confirm that it gets a valid address from the intended scope, with the right gateway and DNS values. If you can capture traffic, you want to see the full exchange: Discover, Offer, Request, and Acknowledgment.

  1. Renew the client lease.
  2. Confirm the address comes from the intended scope.
  3. Check that the gateway and DNS values are right.
  4. If you can capture traffic, verify the full exchange.

A quick packet capture is still the best proof that the fix worked. It shows whether the Discover leaves the VLAN, whether the Offer returns, and whether the final lease handshake completes. If the client now gets a lease but only one device does, keep looking for a port-specific issue instead of assuming the whole VLAN is fixed.

Conclusion

If the VLAN still will not lease after these checks, go back to the last point where traffic was proven to disappear and fix only that layer. If you can’t prove the break point, don’t guess at the server first.

Frequently Asked Questions

What should I check first if only one device in the VLAN fails DHCP?

Check that device’s access port, VLAN assignment, NIC settings, and any port security or filtering first. If every other host on the same VLAN gets a lease, the relay and scope are less likely to be the problem.

Does a static IP working mean DHCP is fixed?

No. A static IP only proves the VLAN can pass normal traffic to the gateway and local network. DHCP can still fail because discovery is broadcast-based and depends on relay, scope availability, and filtering that static traffic does not use.

Sources