Troubleshooting labcore6 steps~16 min5 devices

Fix the desks DHCP forgot

The clinic's desks come up with no address. Find why the DHCP server never hears them — then why their leases still go nowhere.

What you'll be able to do: Both desks lease an address, a mask and a working gateway from a server one subnet away — and you can tell "no answer to the DISCOVER" from "an answer with the wrong options" at a glance.

Open the broken network in the lab

Topics: Troubleshooting · DHCP · DHCP relay · Default gateway

The network you're handed

Clinic-GW — a router, the clinic's router and the desks' gateway
HQ-GW — a router, the head-office router that runs the DHCP pool
SW-Clinic — a switch, the switch the desks plug into
PC-Triage — a pc, the triage desk
PC-Pharmacy — a pc, the pharmacy desk

Step by step

1. Confirm what the desks actually have
The desks "have no network". Look before you guess: `ipconfig /all` on PC-Triage shows whether it is set for DHCP and what it was given. Ask for a lease once more and watch the Console tab for the answer.
On PC-Triage — Read the adapter's settings, then retry the lease
```
ipconfig /all
ipconfig /renew
```
On PC-Pharmacy — Check the second desk the same way
```
ipconfig
```
Check: run ipconfig /all on PC-Triage and look for DHCP Enabled. . . . . . . . . . . : Yes.
Why: No address at all means the host never got an answer to its DHCPDISCOVER — a wrong address would mean it did. `DHCP Enabled: Yes` rules out a host that was simply left on static settings. Both desks failing the same way points at what they share: the path from this LAN to the DHCP server.
2. Follow the DISCOVER to the router
The pool is not in the clinic — it lives on HQ-GW, one subnet away. Check the three things the lease depends on: HQ-GW is reachable from the clinic router, HQ-GW has a pool for 192.168.50.0/24, and Clinic-GW's LAN port is set to relay DHCP toward it.
On Clinic-GW — Reach the DHCP server, then inspect the LAN port the broadcasts arrive on
```
enable
ping 10.99.0.2
show ip interface Gi0/0
```
On HQ-GW — Confirm the pool exists and covers the clinic
```
enable
show ip dhcp pool
```
Check: run show ip interface Gi0/0 on Clinic-GW and look for Helper address is not set.
Why: A DISCOVER is a broadcast from a host with no address yet, and routers never forward broadcasts. The one exception is an `ip helper-address` on the interface the broadcast arrives on: it turns the DISCOVER into a unicast to the named server. The server is reachable and has a pool for this subnet, so the missing relay is the whole reason nobody hears the desks.
3. Relay the broadcasts to HQ-GW, then renew
Add the helper address on the port the desks' broadcasts arrive on — Clinic-GW's Gi0/0 — pointing at the DHCP server. Then ask for a lease again from both desks.
On Clinic-GW — Relay DHCP from the clinic LAN to the server at head office
```
enable
configure terminal
interface Gi0/0
ip helper-address 10.99.0.2
end
```
On PC-Triage — Ask for a lease again
```
ipconfig /renew
```
On PC-Pharmacy — Ask for a lease again
```
ipconfig /renew
```
Check: run show ip dhcp binding on HQ-GW and look for 192.168.50.21.
Why: The relay writes its own LAN address, 192.168.50.1, into the request as the gateway address, and the server uses that to choose the pool for the clinic's subnet — which is how one central server can serve many branches. The first 20 addresses are excluded for fixed devices, so leases start at .21.
4. Leases arrive, but the desks cannot leave the LAN
PC-Triage has an address now, yet it still cannot reach head office. It CAN reach 192.168.50.1. Read the gateway it was given, then read the pool that gave it.
On PC-Triage — Test beyond the LAN, then the gateway, then read the lease
```
ping 10.99.0.2
ping 192.168.50.1
ipconfig
```
On HQ-GW — Read the options the pool hands out
```
enable
show ip dhcp pool
```
Check: run show ip dhcp pool on HQ-GW and look for Default router: 192.168.50.254.
Why: Reaching the gateway's address but nothing beyond it means the host sends off-subnet traffic somewhere else — the same fault as a hand-typed wrong gateway, except this time the wrong value arrived with the lease. 192.168.50.254 belongs to nobody, so the host's ARP for it goes unanswered. The fix belongs in the pool: fixing each PC by hand lasts only until its next renewal.
5. Correct the pool's default router, then renew
Enter the pool again on HQ-GW and set the default router to the address Clinic-GW really owns. The desks keep their old options until they ask again, so renew both.
On HQ-GW — Point the pool's default router at Clinic-GW's LAN address
```
enable
configure terminal
ip dhcp pool CLINIC
default-router 192.168.50.1
end
```
On PC-Triage — Pick up the corrected option
```
ipconfig /renew
```
On PC-Pharmacy — Pick up the corrected option
```
ipconfig /renew
```
Check: run ipconfig on PC-Triage and look for Default Gateway . . . . . . . . : 192.168.50.1.
Why: Typing `default-router` again inside the pool replaces the old value. The server remembers each client by its MAC, so a renewal keeps the same address and simply brings the corrected gateway with it.
6. Prove it from head office
Close the loop from the other end: HQ-GW pings a desk by the address it leased, and the binding table lists both clients.
On HQ-GW — Reach a desk across the WAN, then list the leases
```
enable
ping 192.168.50.22
show ip dhcp binding
```
Check: run show ip dhcp binding on HQ-GW and look for 192.168.50.22.
Why: The binding table is the server's record of who holds what. A lease in the table, a working gateway on the host and a ping in both directions together prove the whole chain: relay, pool, options and routing.

The theory behind it

Fix the broken office — Nobody in a two-desk office can reach the intranet server. Find the two faults and fix them, one layer at a time.
Fix the VLAN that stops at the trunk — Engineering's server is unreachable and the south desk is cut off completely. Find the wrong VLAN and the trunk that drops it.
Fix the NAT that translates nothing — The shop's desks and tills cannot reach the internet, yet the router itself can. Find why nothing leaves translated.
Fix the OSPF adjacencies — Three routers run OSPF and nothing converges. Find the area mismatch and the network statement that matches nothing.
Fix the branch that fails four ways — A branch depot has lost head office, and two of its hosts have faults of their own. Find all four, one layer at a time.

Build it for real

The lab walks you through these steps and ticks each one off as your network starts working.

Open in the lab

The network you're handed

Step by step

1. Confirm what the desks actually have

2. Follow the DISCOVER to the router

3. Relay the broadcasts to HQ-GW, then renew

4. Leases arrive, but the desks cannot leave the LAN

5. Correct the pool's default router, then renew

6. Prove it from head office

The theory behind it

More in Troubleshooting

Build it for real