Full syllabusWeek 6Day 29

Wireless Access

Wired and wireless client troubleshooting

You will collect comparable Windows, macOS and Linux evidence, recognize missing or wrong IP parameters, separate RF and security symptoms from DHCP and DNS questions, and retest a controlled client disconnect without changing infrastructure.

Exam map
Current v1.1
1.10-1.11, 2.6-2.9
Announced v2.0
1.6
Domain
Network access and connectivity

First, the essential ideas

Adapter and interface
An adapter connects a host to a network. A host may have Ethernet, Wi-Fi, VPN and virtual interfaces at once; inspect the one actually carrying the test traffic.
Association and security
Association connects a wireless station to an AP's radio service. WPA authentication and key establishment must also complete before protected user traffic can flow.
SSID and BSSID
SSID is the wireless service name. BSSID identifies the particular radio service. Record both where permitted, because the same SSID may be offered by several APs.
IP address, prefix and gateway
The address identifies an interface and its prefix defines the local subnet. The default gateway is the local router used for destinations without a more specific route.
DHCP and lease
DHCP provides a timed address assignment and related options. A lease is not proof that today's wireless association or application is healthy; cached information can remain.
APIPA and link-local
Windows APIPA is automatic IPv4 link-local addressing in 169.254.0.0/16. Such addresses are for local-link use, not ordinary routed reachability.
DNS and resolver
DNS translates names into records such as IP addresses. A resolver performs lookups. An IP test and a name lookup answer different questions.
Ping and ICMP
Ping uses ICMP echo exchanges to test one IP destination. No reply can reflect filters, rate limits or either path direction; it is not a complete diagnosis.
RSSI, SNR and interference
RSSI is received signal-strength indication. Signal-to-noise ratio (SNR) compares signal with noise. Interference, contention and client behavior can affect service even with strong signal.
Baseline and rollback
A baseline records known settings and actual tests before a change. Rollback restores the specific changed setting and repeats the original test, rather than resetting unrelated network state.

Choose one client, one interface and one destination before running a command. Compare the observation with the address plan, then ask the next narrow question. The lab's three OS tracks collect equivalent facts using different interfaces; unavailable platforms remain explicitly untested.

See the idea

Find the first unsupported checkpoint

A later-stage symptom does not erase the need to verify earlier client evidence.

A diagnostic ladder separates link, security, IP settings and destination tests. Each record asks a different question about the same client.Link orassociationCorrect activeadapterWireless securityCompatiblecredentialsIP parametersAddress and next hopDestination testIP then name service
A diagnostic ladder separates link, security, IP settings and destination tests. Each record asks a different question about the same client.Link orassociationCorrect activeadapterWireless securityCompatiblecredentialsIP parametersAddress and next hopDestination testIP then name service
Step 1 of 3

Check access first

Read physical link or current wireless association, then verify the intended security method.

What this model leaves out: Conceptual evidence ladder, not a physical topology. Passing one checkpoint does not prove later services; RF behavior requires a real authorized radio observation.

Full visual explanation and references

A diagnostic ladder separates link, security, IP settings and destination tests. Each record asks a different question about the same client.

Conceptual evidence ladder, not a physical topology. Passing one checkpoint does not prove later services; RF behavior requires a real authorized radio observation.

  1. Check access first

    Read physical link or current wireless association, then verify the intended security method.

    www.cisco.com
  2. Compare the plan

    Inspect address, prefix and gateway on the active interface before blaming an upstream device.

    www.cisco.comwww.rfc-editor.org
  3. Scope the conclusion

    Test one intended IP destination. Diagnose DNS separately if the required name service exists.

    www.cisco.com

One idea at a time

Start with something familiar.

Start by writing the client's interface name, current connection state and intended destination. Leave blank any field you cannot actually observe. A missing OS or Wi-Fi adapter is a limitation to record, not an invitation to invent output.

Why learn this?

A reset can remove useful evidence and interrupt unrelated services. A staged check helps distinguish an incorrect address, failed key exchange or absent DNS server before any change is made.

An everyday comparison

When a letter does not arrive, checking the sender's address, local collection point and recipient address answers different questions. A correctly written envelope alone does not show delivery.

What happens in a network

The address plan resembles the labels, the gateway resembles the next delivery point, and a destination test checks a particular exchange. Each observation narrows the question.

Where the comparison stops

The postal comparison cannot explain wireless authentication, IP routing or DNS behavior. A network can filter replies, and failed ping is not proof that a destination is powered off.

Follow one worked example

  1. Choose the actual traffic adapter.

    Find Wi-Fi, Ethernet or a virtual interface and associate its current address with the intended test.

    Why: A valid address on an unused adapter can distract from a disconnected or misconfigured active one.

  2. Collect the same facts on each OS.

    Record link state, address/prefix, gateway, DNS and wireless security using each platform's documented inspection surface.

    Why: Comparable columns make different command output formats easier to interpret.

  3. Compare a controlled disconnect.

    After a working local test, disable Wi-Fi only on the owned Windows test client, observe the change, restore it and repeat the same target.

    Why: This isolates association loss without changing the AP, shared key or router configuration.

Your first small task

Given a client at 198.51.100.10/24, gateway 198.51.100.1 and destination 198.51.100.20, write whether the destination is local. Then explain why reaching it would not test the default gateway's remote forwarding.

Show a hint

Compare the first 24 bits, which are the first three octets here, before deciding whether a gateway is required.

Check the expected result

The destination is in the same /24 subnet. Its local exchange does not establish the router's ability to forward toward a remote network.

Pause and explain it in your own words

Wi-Fi shows connected and ipconfig displays 169.254.8.9. Is an ordinary routed IPv4 service established?

Show a hint

Separate association from the source and scope of the IPv4 address.

Show the explanation

No. Association does not prove that DHCP succeeded, and IPv4 link-local addressing does not provide ordinary routed access. Check whether DHCP was intended, then inspect the lease path and service VLAN without blaming radio strength alone.

Short answer

What should you understand today?

Troubleshoot the active client interface in stages: physical link or wireless association, security, IP settings, local gateway, remote destination and name resolution. Record what each check proves. A strong Wi-Fi signal or an address displayed by the OS does not establish current authentication, correct routing or a healthy application.

Before you begin

Learning targets

Prerequisites
  • Use owned test clients or instructor-provided screenshots; do not collect other people's wireless traffic.
  • Know IPv4 masks and default gateways; this lesson repeats the relevant meanings.
  • The practical track uses the prepared isolated WLAN from Day 28, recreated by its instructor; no Internet or DNS service is provided.
  • Record operating-system versions and actual interface names; Linux commands require NetworkManager.
By the end
  • Inspect address, prefix, gateway and resolver information on all three client OS families.
  • Identify active wired versus wireless interface evidence and security parameters.
  • Separate RF, authentication, DHCP, DNS and routing hypotheses.
  • Perform one client-only disconnect and restore the original association.
  • State what remains unmeasured when a platform or physical radio is unavailable.

Words you will meet

Open a word when you need its meaning. You do not need to memorize the whole list before reading.

Adapter and interface

An adapter connects a host to a network. A host may have Ethernet, Wi-Fi, VPN and virtual interfaces at once; inspect the one actually carrying the test traffic.

Association and security

Association connects a wireless station to an AP's radio service. WPA authentication and key establishment must also complete before protected user traffic can flow.

SSID and BSSID

SSID is the wireless service name. BSSID identifies the particular radio service. Record both where permitted, because the same SSID may be offered by several APs.

IP address, prefix and gateway

The address identifies an interface and its prefix defines the local subnet. The default gateway is the local router used for destinations without a more specific route.

DHCP and lease

DHCP provides a timed address assignment and related options. A lease is not proof that today's wireless association or application is healthy; cached information can remain.

APIPA and link-local

Windows APIPA is automatic IPv4 link-local addressing in 169.254.0.0/16. Such addresses are for local-link use, not ordinary routed reachability.

DNS and resolver

DNS translates names into records such as IP addresses. A resolver performs lookups. An IP test and a name lookup answer different questions.

Ping and ICMP

Ping uses ICMP echo exchanges to test one IP destination. No reply can reflect filters, rate limits or either path direction; it is not a complete diagnosis.

RSSI, SNR and interference

RSSI is received signal-strength indication. Signal-to-noise ratio (SNR) compares signal with noise. Interference, contention and client behavior can affect service even with strong signal.

Baseline and rollback

A baseline records known settings and actual tests before a change. Rollback restores the specific changed setting and repeats the original test, rather than resetting unrelated network state.

01

Start with link and association evidence

For a wired client, record cable connection, adapter state, negotiated speed and errors available from the OS or switch. For a wireless client, record SSID, BSSID where permitted, connection state and negotiated security. Do not treat the saved Wi-Fi profile as the current association. Physical distance, obstacles, channel use and noise require physical observations; no virtual cable in GNS3 or CML can reproduce those conditions.

  • A signal-strength indicator is not an application-health measurement.
  • Inspect only authorized devices and do not inject interference to create a fault.
02

Compare client parameters across Windows, macOS and Linux

The evidence columns are the same across systems even when their displays differ: interface, address, prefix or mask, gateway, DNS and assignment method. Windows ipconfig /all displays adapter details; netsh wlan show interfaces adds current wireless information. macOS exposes the selected service through System Settings > Network and Wi-Fi Details, including TCP/IP and DNS. Linux with NetworkManager provides device and connection inspection through nmcli. Official OS links in the setup notes support the exact client procedure; Cisco's source supports the troubleshooting method.

  • Do not copy another operating system's syntax into a client console.
  • Record unavailable fields and permissions rather than trying to bypass them.
03

An address must match the intended service

Check whether the client should use static addressing or DHCP. A DHCP-enabled client with a link-local address warrants investigation of assignment, but one observation does not identify the failed server or relay. A correct-looking address can be stale, duplicated or assigned to an inactive interface. A wrong mask can make a remote destination appear local; a wrong gateway can break remote paths while same-subnet peers still work. IPv6 must be assessed separately rather than assumed broken because IPv4 is wrong.

  • A client without a DHCP lease is not necessarily faulty when static addressing was intended.
  • IPv4 link-local addresses are not a substitute for a planned routed address.
04

Use reachability tests with narrow conclusions

Test a documented local destination before an approved remote destination, and compare replies with interface and route evidence. A self-ping tests local stack behavior, not a physical cable or AP path. A gateway reply confirms that particular exchange but not the gateway's upstream route. Missing replies can reflect filters, host settings, load or the return path. Preserve outputs and timestamps so the next operator can distinguish persistent failure from transient neighbor discovery.

  • Keep the source interface and destination fixed when comparing a fault and repair.
  • Do not declare every protocol healthy from one successful ICMP exchange.
05

Separate DNS and wireless security from IP routing

A name lookup asks a resolver for records. If an IP endpoint works but its name fails, inspect the configured resolver, suffix and returned record before flushing caches. A correct DNS answer can still point to a service that is unreachable or unavailable. Wireless security failures occur earlier: mismatched WPA mode, unsupported cipher, wrong PSK or Enterprise certificate validation can prevent data service. Do not disable certificate checking, authentication or endpoint firewalls simply to obtain a green test.

  • An unavailable service is different from a misconfigured service.
  • Keep passwords, private keys and full client identifiers out of shared diagnostic notes.
06

Restore one change and report the remaining uncertainty

The deliberate fault is turning Wi-Fi off on the owned Windows test laptop after a working baseline. The AP, VLAN and credentials remain unchanged. Restore Wi-Fi, select the same test network and repeat the original destination. This demonstrates a local access dependency when actually executed; it does not measure roaming, interference or throughput. For macOS and Linux, collect equivalent inspection evidence without requiring a second fault.

  • A paper interpretation must be labeled predicted rather than observed.
  • Record one platform as unavailable rather than marking all three client tracks complete.

Real-world walkthrough

The no-Internet icon was not the local fault

In this hypothetical training rack, a Windows learner sees 'No Internet' and assumes the WLAN password is wrong. The rack has no Internet connection by design, while the real local test is a ping to Router1. Meanwhile a Linux client uses a different interface than expected. Both cases require a scoped evidence ledger rather than a blanket network reset.

  1. Confirm the test requirement is local Router1 reachability, not public web access or DNS resolution.
  2. Identify the active adapter and collect its address, prefix, gateway and assignment method on the actual OS.
  3. Record wireless association and negotiated security independently of any browser connectivity icon.
  4. Run the same intended destination test and compare with a known-working client on the same service.
  5. Correct only an evidenced setting, then repeat the original test and state which services were never in scope.
Diagnostic tools become useful when tied to an expected service and an actual interface. A client can be locally healthy without Internet access, and a displayed address can belong to the wrong adapter.

Guided GNS3 lab

Three client inspection tracks and one reversible disconnect

Collect equivalent Windows, macOS and Linux observations on an instructor-prepared isolated network. Run only on available owned clients, label missing tracks untested, and compare baseline, client Wi-Fi-off fault and restoration on Windows.

Use WLC1/AP1/SW1/Router1 prepared as Day28. AP1 Ethernet0 -> SW1 Gi1/0/1 VLAN10; WLC1 port1 -> Gi1/0/24 trunk VLAN10,20; Router1 Gi0/0 -> Gi1/0/3 VLAN20. Each OS client joins QCS-LAB28; for optional wired comparison, its Ethernet0 replaces Wi-Fi and connects to Gi1/0/4 access VLAN20.
Prepared AireOS 8.10.190.0 WLC3504 and AP1832i Local mode; no infrastructure changes.Win1 Windows11, Mac1 macOS15, Linux1 Ubuntu24.04 Desktop with NetworkManager; physical Wi-Fi adapters required.Router1 local responder and SW1 prepared PoE/VLAN switch; use one available OS track at a time if necessary.

Addressing plan

DeviceInterfaceAddressPurpose
Router1Gi0/0198.51.100.1/24Local target and client gateway
Win1Wi-Fi or Ethernet198.51.100.10/24Static; gateway 198.51.100.1; DNS absent
Mac1Selected service198.51.100.11/24Static; router 198.51.100.1; DNS absent
Linux1Discovered adapter198.51.100.12/24Static; gateway 198.51.100.1; DNS absent
WLC1management / lab20192.0.2.2/24 / 198.51.100.2/24Controller interfaces prepared by instructor
AP1management192.0.2.11/24Prepared AP join

Set up the lab

  1. Instructor recreates Day28's WPA2-AES QCS-LAB28 service, confirms local reachability and privately supplies its disposable credential. No DHCP, DNS or Internet is present.
  2. Record previous client settings. Set the listed static parameters using each OS Network settings GUI. Leave unrelated managed clients untouched; a single test adapter must carry each observation.
  3. Windows command references: https://learn.microsoft.com/en-us/windows-server/administration/windows-commands/ipconfig and https://learn.microsoft.com/en-us/windows-server/administration/windows-commands/netsh-wlan. Never export profiles with plaintext keys.
  4. macOS inspection reference: https://support.apple.com/en-gb/guide/mac-help/mh11395/mac. Use System Settings > Network > selected service > Details, TCP/IP and DNS.
  5. Linux reference: https://networkmanager.pages.freedesktop.org/NetworkManager/NetworkManager/nmcli.html. Confirm NetworkManager manages the chosen adapter; do not install or reconfigure network services during this exercise.
  6. If any OS or radio is unavailable, use a labeled worksheet for that track. A VM can inspect virtual IP settings but does not establish physical wireless behavior.
01

Identify links and the test boundary

On the rack worksheet, mark every port above and record whether each OS track is real or paper-only. Verify only owned clients join the test SSID. Record the absence of DHCP, DNS and Internet so later errors are not mistaken for new faults.

What you should see
The evidence ledger identifies the actual clients, available platforms and exact local test target.
Why this step matters
A clear service boundary keeps a generic 'no Internet' indicator from overriding the actual local test requirement.
02

Inspect Windows adapter and wireless state

On Win1, open Start > Command Prompt before typing. Use the Wi-Fi adapter's output, not a disconnected virtual adapter. If wireless details are permission-restricted, use the authorized Windows Wi-Fi properties view and record the limitation.

ipconfig /all
netsh wlan show interfaces
ping 198.51.100.1

What each command means

ipconfig /all
Display full adapter addressing, mask, gateway, DNS and assignment information; inspect the active Wi-Fi entry.
netsh wlan show interfaces
Display detected wireless interfaces and available current association/security details without revealing a stored key.
ping 198.51.100.1
Test the planned local router from Win1 using ICMP; preserve the actual bounded result.
What you should see
Expect the planned static settings, QCS-LAB28 association and local replies when the prepared rack is working.
Why this step matters
This correlates current radio association with addressing and one specific service path.
03

Inspect macOS parameters and security

On Mac1 open System Settings > Network, select Wi-Fi and open Details. Record TCP/IP address, mask and Router plus DNS entries; inspect Wi-Fi's current network/security details. Option-click the Wi-Fi menu and open Wireless Diagnostics if authorized; do not change the service.

What you should see
Expect 198.51.100.11/24, Router 198.51.100.1 and the prepared WPA2 service; DNS is intentionally absent.
Why this step matters
The graphical view provides platform-appropriate inspection without relying on retired private wireless command paths.
04

Test the macOS local exchange

On Mac1 open Applications > Utilities > Terminal before typing. Confirm the selected Wi-Fi service is active and other data paths are not carrying the test. Do not treat a local self-ping as a substitute for this router destination.

ping -c 4 198.51.100.1

What each command means

ping -c 4 198.51.100.1
Send four ICMP probes to Router1 from Mac1; -c limits the count so the test ends without a manual interruption.
What you should see
Record the actual four-probe result; replies support only this local IP exchange.
Why this step matters
The same destination makes the Mac evidence comparable with the Windows baseline despite different command syntax.
05

Inspect the Linux NetworkManager state

On Linux1 open its desktop Terminal before typing. Confirm NetworkManager is installed and manages the test adapter; otherwise stop this CLI track. Read the adapter name from device status and correlate it with the IP information.

nmcli device status
nmcli device show
nmcli connection show --active
nmcli device wifi list
ping -c 4 198.51.100.1

What each command means

nmcli device status
List interface names, types and connection state so the actual Wi-Fi adapter can be identified.
nmcli device show
Display per-device properties including addresses, gateway and DNS; inspect the identified test adapter.
nmcli connection show --active
List active connection profiles; a saved inactive profile is not the active network.
nmcli device wifi list
List visible Wi-Fi service information and the active marker; this does not reveal the private PSK or prove service health.
ping -c 4 198.51.100.1
Send four ICMP probes to the same local Router1 target and retain the observed result.
What you should see
Expect the Linux test adapter at 198.51.100.12/24 and the intended profile, then local replies if the path works.
Why this step matters
This collects the same categories as the other OS tracks while respecting Linux's installed network manager.
06

Compare a wired observation

For one owned client only, record its Wi-Fi baseline, turn Wi-Fi off, and connect its Ethernet port to SW1 Gi1/0/4 access VLAN20. Apply that client's same planned static address to Ethernet only. Inspect the selected OS network settings, then restore the original Wi-Fi arrangement.

What you should see
The wired interface has a physical link and its own configuration; wireless SSID and BSSID are not wired parameters.
Why this step matters
A wired comparison can narrow an RF-specific hypothesis, but a different VLAN or address would make the comparison misleading.
07

Create the controlled Windows disconnect

After restoring Win1's original Wi-Fi baseline, use Windows Quick Settings to turn Wi-Fi off on Win1 only. Keep its Ethernet unplugged. Observe adapter state; do not change WLC, AP, client credentials or gateway configuration.

What you should see
Win1 loses its wireless data path; a retained static address does not establish connection.
Why this step matters
This intentionally changes association availability without injecting RF interference or disrupting infrastructure shared by other clients.
08

Record the failed same-target test

On Win1 open Start > Command Prompt before typing. Confirm Wi-Fi is off and no alternative data adapter is active. Record both the adapter display and the router test; error wording may vary with OS state.

ipconfig /all
ping 198.51.100.1

What each command means

ipconfig /all
Observe the adapter's disconnected state and any retained configuration instead of assuming the address disappeared.
ping 198.51.100.1
Repeat the original Router1 test during the controlled disconnect; no successful exchange is expected over the disabled path.
What you should see
Expect no successful Router1 exchange from Win1, with actual output retained rather than invented wording.
Why this step matters
The observation links the controlled access fault to the same destination used in the working baseline.
09

Restore and verify the original connection

On Win1 turn Wi-Fi on, select QCS-LAB28 and verify the planned static settings. Then open Start > Command Prompt before typing. Keep the same local destination and compare timestamps with the baseline and fault.

netsh wlan show interfaces
ipconfig /all
ping 198.51.100.1

What each command means

netsh wlan show interfaces
Confirm the intended association is restored where Windows permits this display.
ipconfig /all
Recheck the active Wi-Fi address, mask and gateway against the original plan.
ping 198.51.100.1
Repeat the same local router test after restoration; retain unexpected failures for further diagnosis.
What you should see
Expected recovery is the original association, address plan and local replies, not a claim of Internet or DNS access.
Why this step matters
Restoration is complete only when the original scoped checks are repeated, not merely when the Wi-Fi toggle is on.

Verify

  • Complete separate evidence columns for Windows, macOS and Linux; label any unavailable track untested.
  • Confirm only the selected adapter carried each test and no self-ping was substituted.
  • Retain baseline, Wi-Fi-off failure and restoration observations from the same Windows destination.
  • Record DNS and Internet as intentionally unprovided, not tested failures.
  • Keep RF coverage, interference, roaming and throughput marked unmeasured.

Troubleshoot

  • If one client fails while another works, compare its actual prefix, gateway and active interface before resetting the AP.
  • If nmcli is absent or an adapter is unmanaged, stop the Linux CLI track and use the documented desktop settings; do not replace its network manager.
  • If macOS or Windows withholds wireless identifiers, respect privacy or management policy and record the missing evidence.
  • If pings fail even before the deliberate fault, investigate the prepared VLAN and responder or filters; never claim the later disconnect caused an already-failing baseline.
Licensing and cleanup

GNS3 does not provide Cisco images. Use software only when the applicable license or entitlement permits that environment. Do not share or redistribute image files. Cisco Modeling Labs is the official alternative; its reference images are licensed for use within CML unless separately licensed. A paper exercise needs no image, but does not establish hands-on competence. Virtual routers do not reproduce physical radios, PoE delivery, hardware forwarding rates or failover timing.

  • Restore each owned client's previous settings and remove only the temporary test wireless profile.
  • Return the rack to its instructor-approved state; do not publish credentials, full client identifiers or diagnostic archives.

Practice set

Answer before you reveal

01Why can a same-subnet test work with the wrong gateway?
Local destinations do not normally need the default gateway.

The host uses its prefix to decide whether a destination is local. A wrong gateway becomes relevant to routed destinations, so choose tests that exercise the suspected dependency.

02Does 169.254.8.9 prove that the DHCP server is down?
No; it indicates link-local addressing, not the exact cause of missing intended assignment.

The client may not reach DHCP because of association, VLAN, relay or server problems. First establish whether DHCP was intended at all.

03What fields should the three OS tracks have in common?
Interface state, address/prefix, gateway, DNS, assignment method and applicable wireless security.

Different tools format the information differently. A comparable ledger records the meaning of each field instead of requiring identical command output.

04Why avoid a blanket DNS cache flush first?
It changes evidence before confirming whether name resolution is the failing stage.

A numeric IP test and a lookup through the configured resolver can isolate the problem. No resolver is supplied in this lab, so flushing a cache cannot create one.

05Does a VM with a virtual Ethernet adapter test RF coverage?
No; physical radio, antenna and environmental behavior are outside that virtual interface.

A VM can provide useful IP configuration practice, but its Ethernet link does not measure channel contention, signal strength or wireless roaming.

06What should follow restoring Wi-Fi on Win1?
Confirm the intended association and static settings, then repeat the original Router1 test.

The toggle only enables the adapter. The client must reconnect to the correct service and successfully exchange traffic on the original path.

Knowledge check

Quiz: prove the reasoning

0/5 answered
01Which observation most directly checks current wireless association?
02Which command inspects Windows adapter addressing?
03Where does this lesson inspect macOS IP parameters?
04What is the purpose of nmcli device status?
05Which result is expected in the isolated rack?

Keep these

Technical takeaways

  • Inspect the interface actually carrying the traffic.
  • Keep association, IP configuration and name resolution as separate checks.
  • Compare equivalent evidence across operating systems, not identical syntax.
  • A local reply does not prove upstream forwarding or application health.
  • Use only a bounded owned-client fault with a recorded restoration.
  • Treat physical RF and unavailable OS tracks as explicitly unmeasured.