<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Will Edwards on Medium]]></title>
        <description><![CDATA[Stories by Will Edwards on Medium]]></description>
        <link>https://medium.com/@willedwardsjr79?source=rss-f4eceb10dfa9------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*Db9AufbT3-iMz9BMG6MQgw.jpeg</url>
            <title>Stories by Will Edwards on Medium</title>
            <link>https://medium.com/@willedwardsjr79?source=rss-f4eceb10dfa9------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sun, 22 Nov 2026 12:07:24 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@willedwardsjr79/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[How a Software Update Became a Spy Network: The SolarWinds Attack Through MITRE ATT&CK]]></title>
            <link>https://medium.com/@willedwardsjr79/how-a-software-update-became-a-spy-network-the-solarwinds-attack-through-mitre-att-ck-bba65ff49d72?source=rss-f4eceb10dfa9------2</link>
            <guid isPermaLink="false">https://medium.com/p/bba65ff49d72</guid>
            <dc:creator><![CDATA[Will Edwards]]></dc:creator>
            <pubDate>Fri, 23 Oct 2026 20:45:26 GMT</pubDate>
            <atom:updated>2026-10-23T20:45:26.957Z</atom:updated>
            <content:encoded><![CDATA[<p><em>A trusted update, a validly signed backdoor, and forged identity tokens: the SolarWinds compromise, explained step by step with MITRE ATT&amp;CK.</em></p><p>The SolarWinds compromise ran for more than a year before anyone found it. MITRE first sees the campaign in August 2019, and the activity was discovered in mid-December 2020.</p><p>The group behind it, APT29, was attributed by the US and UK governments in April 2021 to Russia’s Foreign Intelligence Service (SVR). It inserted malicious code into the SolarWinds Orion build process, so a normal software update delivered the backdoor to roughly 18,000 customers. A much smaller number were then chosen for follow-on access.</p><p>Here is the attack in five steps, with the MITRE ATT&amp;CK technique behind each one. The details come from MITRE’s campaign page for C0024, which lists more than 60 techniques, and its page for the SUNBURST malware; these eleven show the arc from supply chain to cloud.</p><h3>The attack in five steps</h3><ol><li><strong>Poison the build.</strong> APT29 got malicious code into the Orion build process, so a routine software update carried the backdoor (T1195.002, Compromise Software Supply Chain). Because the build process signed it, SUNBURST arrived with a valid SolarWinds signature (T1553.002, Code Signing).</li><li><strong>Blend in.</strong> Software and DLLs were renamed to look benign (T1036.005, Match Legitimate Resource Name or Location). SUNBURST remained dormant after initial access for up to two weeks (T1497.003, Time Based Checks), a behavior MITRE records on its SUNBURST software page. Command and control looked like ordinary web traffic (T1071.001, Web Protocols) and used dynamically generated subdomains (T1568, Dynamic Resolution).</li><li><strong>Use real credentials.</strong> Compromised credentials drove remote access and lateral movement (T1078, Valid Accounts). The attackers also added credentials to OAuth apps and service principals to keep their access (T1098.001, Additional Cloud Credentials).</li><li><strong>Forge the keys.</strong> Using compromised SAML signing certificates, they forged tokens (T1606.002, Forge Web Credentials: SAML Tokens). They also changed the federation trust so the environment accepted tokens signed by their own certificate (T1484.002, Trust Modification).</li><li><strong>Read the mail.</strong> Mailboxes of executives and IT staff were exported (T1114.002, Remote Email Collection).</li></ol><h3>Where defenders could have seen it</h3><p>This section and the next are analyst judgment, not MITRE findings. Detection was weakest at the start of the chain and most realistic near the end.</p><ul><li><strong>Start of the chain.</strong> Signed, trusted software behaving badly gives almost no endpoint signal. Build integrity checks and reproducible builds are the real controls.</li><li><strong>Middle of the chain.</strong> Stolen credentials and forged tokens look like ordinary logins unless identity logs are correlated with what the tokens were used for.</li><li><strong>End of the chain.</strong> Reading mailboxes through cloud APIs leaves audit records, but only where access by application permission is logged and reviewed.</li></ul><h3>What to alert on</h3><p>The strongest detections sit in identity and DNS telemetry, because the endpoint looked legitimate.</p><ul><li><strong>DNS (T1071, T1568).</strong> Unusual subdomain entropy, and Orion servers talking to unfamiliar external domains.</li><li><strong>New credentials on service principals (T1098.001).</strong> Especially outside change windows.</li><li><strong>Tokens with no matching login (T1606.002).</strong> Compare token issuance against actual authentication events; a token with no login behind it is a strong signal.</li><li><strong>Federation trust changes (T1484.002).</strong> Alert on any change.</li><li><strong>Mailbox access by app permission (T1114.002).</strong> Audit access by application permission, not just by user.</li></ul><h3>Defenses from MITRE D3FEND</h3><p>MITRE’s D3FEND matrix is the defensive counterpart to ATT&amp;CK. Three of its countermeasures line up with this attack:</p><ul><li><strong>Command and control patterns:</strong> Network Traffic Analysis and DNS Traffic Analysis</li><li><strong>Identity:</strong> Credential Hardening and Authentication Event Thresholding</li><li><strong>Supply chain:</strong> Software Update and Platform Monitoring</li></ul><h3>The takeaway</h3><p><strong>Detection effort belongs in identity and cloud telemetry, because that is where the attacker was observable once trusted software had given them a foothold.</strong> A write-up that maps where each stage could realistically be seen, not just which techniques were used, is what separates a threat report from a technique list.</p><h3>Sources and caveats</h3><p>Technique IDs and names follow MITRE’s C0024 page as read on October 15, 2026. ATT&amp;CK is renumbered from time to time, and in v20 the techniques once grouped under Defense Evasion are listed under Stealth or Defense Impairment, so recheck IDs, names and tactic labels before you publish. T1497.003 comes from MITRE’s SUNBURST malware page rather than the campaign page, the D3FEND names were not rechecked, and the CISA advisory was not read for this write-up.</p><ul><li><a href="https://attack.mitre.org/campaigns/C0024/">MITRE ATT&amp;CK, SolarWinds Compromise (C0024)</a></li><li><a href="https://attack.mitre.org/software/S0559/">MITRE ATT&amp;CK, SUNBURST (S0559)</a></li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=bba65ff49d72" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Simulated Data Breach Incident Report]]></title>
            <link>https://medium.com/@willedwardsjr79/simulated-data-breach-incident-report-ff74acd2c572?source=rss-f4eceb10dfa9------2</link>
            <guid isPermaLink="false">https://medium.com/p/ff74acd2c572</guid>
            <category><![CDATA[incident-management]]></category>
            <category><![CDATA[incident-reporting]]></category>
            <category><![CDATA[data-breach-report]]></category>
            <category><![CDATA[cybersecurity]]></category>
            <dc:creator><![CDATA[Will Edwards]]></dc:creator>
            <pubDate>Fri, 23 Oct 2026 06:01:19 GMT</pubDate>
            <atom:updated>2026-10-23T06:01:19.708Z</atom:updated>
            <content:encoded><![CDATA[<p><strong>Orion Health Services<br></strong>Cyber Incident Report<br>Analyst: Will Edwards</p><p><strong>Executive Summary</strong></p><p>Orion Health Services suﬀered a ransomware attack the began with a phishing email to a finance team member that ended with encrypted files, a ransom note on the shared drive and exposure of sensitive data. Investigation is ongoing; this report reflects what is confirmed as of October 22, 2026</p><ul><li><strong>What Happened: </strong>A malicious Excel attachment gave the attacker initial access. The attacker then harvested credentials with Mimikatz, logged in from an overseas IP and encrypted the file with the .orionlock extension.</li><li><strong>Detection:</strong> The IT team noticed unusual outbound traﬃc from an internal server Monday morning. By midday, staﬀ were locked out of systems and a ransom note appeared</li><li><strong>Data at risk: </strong>Employee payroll records, patient appointment schedules and internal system credentials.</li><li><strong>Systems aﬀected: </strong>File server, HR and finance systems and the backup server (partially encrypted).</li></ul><p><strong>Decisions needed from the executive team:</strong> approve external incident response and legal support, decide on regulator and customer notification, and confirm the position on any ransom demand (this report recommends against paying without legal and law-enforcement advise.)</p><h3><strong>Incident Overview</strong></h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/997/1*v48fJH8SSXOM9KwwTfCl_Q.png" /></figure><h3><strong>Timeline of Events</strong></h3><p>Only Monday’s detection and lockouts are timed in the source information; earlier steps are placed in their likely order and need log evidence to confirm.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/992/1*SftuC3DS2rqc9MUAeaIfCA.png" /></figure><h3><strong>Attack Chain Analysis</strong></h3><p>The evidence fits a standard ransomware sequence. Steps 1 to 3 and 6 are supported by the stated IOCs; steps 4 and 5 are inferences that need confirmation.</p><ol><li><strong>Initial access (confirmed entry point):</strong> A finance team member received a phishing email with a malicious Excel attachment. Opening it most likely ran embedded code (for example, a macro) that gave the attacker a foothold.</li><li><strong>Credential harvesting (confirmed IOC):</strong> The attacker used Mimikatz, which extracts passwords and tokens from memory. This is how internal system credentials were compromised.</li><li><strong>Unauthorized access (confirmed IOC): </strong>A suspicious login from an overseas IP suggests the attack used stolen credentials to work remotely.</li><li><strong>Lateral movement (inferred):</strong> Reaching the file server, HR and finance systems and the backup server implies the attacker moved across the network with valid credentials.</li><li><strong>Data exfiltration (suspected):</strong> The unusual outbound traﬃc from an internal server on Monday morning is consistent with data theft before encryption. Volume and destination must be confirmed from network logs.</li><li><strong>Encryption and extortion (confirmed):</strong> Files were encrypted with</li><li>the .orionlock extension, users were locked out, and a ransom note appeared on the shared drive. Encrypting the backup server, even partially, reduces Orion’s ability to restore quickly.</li></ol><h3><strong>Indicators of Compromise</strong></h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1001/1*Ql7KQUBFQxhHJhJLFBRtDg.png" /></figure><h3><strong>Regulatory and Legal Considerations</strong></h3><p>Orion operates in two jurisdictions, so both breach-notification regimes need review by legal counsel.</p><ul><li><strong>Australia:</strong> The Notifiable Data Breaches scheme under the Privacy Act 1988 generally requires notifying the OAIC and aﬀected individuals when a breach is likely to cause serious harm.</li><li><strong>New Zealand:</strong> The Privacy of 2020 requires notifying the Privacy Commissioner and aﬀected people about breaches likely to cause harm as soon as practical.</li><li><strong>Clinic contracts: </strong>Orion’s agreements with clinics may require notifying them within set periods, and clinics may have their own duties to patients.</li></ul><p>This is a security analysis, not legal advice; counsel should confirm obligations and deadlines. The clock for assessment and notification may already be running from when Orion became aware of the breach on Monday.</p><h3><strong>Containment, Eradication, and Recovery</strong></h3><p>The source information does not say which response actions are already done, so these are recommended action. Tick oﬀ each as IT confirms it.</p><p><em>Immediate containment</em></p><ul><li>Isolate aﬀected servers and the finance team member’s device from the network (do not power oﬀ, to preserve memory evidence).</li><li>Disable the compromised accounts and block the overseas IP at the firewall.</li><li>Disconnect the backup server and any clean backups from the network.</li><li>Block the malicious attachment and sender across the email gateway.</li><li>Preserve evidence: disk images, memory captures, logs, and a copy of the ransom note.</li></ul><p><em>Eradication</em></p><ul><li>Reset all passwords, including service and admin accounts, and reset the Kerberos KRBTGT account twice.</li><li>Hunt for Mimikatz, persistence mechanisms, and other attacker tools on all endpoints and servers.</li><li>Rebuild aﬀected systems from known-good images rather than cleaning them in place.</li></ul><p><em>Recovery</em></p><ul><li>Verify which backups are intact and scan them for malware before any restore.</li><li>Restore in priority order: identity systems, HR and finance, file server, then the rest.</li><li>Monitor closely for reinfection and repeat logins from the attacker’s infrastructure.</li></ul><p><em>Communications and External Support</em></p><ul><li>Engage legal counsel and an external incident response firm.</li><li>Notify law enforcement and the relevant cyber agencies in Australia and New Zealand.</li><li>Decide on regulator, clinic, and staﬀ notifications (see impact assessment).</li><li>Hold the ransom position: no payment or contact with the attacker without legal and law enforcement advice.</li></ul><h3><strong>Root Cause and Contributing Factors</strong></h3><p>The root cause was a successful phishing email that delivered a malicious Excel file to a financial team member. The factors below are hypotheses based on the attack pattern and should be checked against configuration logs.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/997/1*C9-SX-VQn15RXz24GaPgxw.png" /></figure><h3><strong>Recommendations and Lessons Learned</strong></h3><p>These recommendations address each stage of the attack so that no single failure can repeat this outcome.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1019/1*FKCJSoCFYKXcNXxaAZUFYg.png" /></figure><p><strong>Lessons learned: </strong>The attacker needed only one opened attachment. Resilience depends on layered controls and tested backups and on a rehearsed plan for notifying regulators and clinics, since the data involved is health-related.</p><h3><strong>Open Questions, Assumptions, and Next Steps</strong></h3><p>This is a preliminary report. The questions below decide the final scope and the notification decisions. Tick oﬀ each one as they are answered.</p><p><em>Open Questions</em></p><ul><li>How much data left the network, to where, and when did the transfer start?</li><li>Which patient records are in the appointment schedules (names, contact details, clinical details), and how many people and clinics are aﬀected?</li><li>Was the stolen data limited to the three listed categories?</li><li>When exactly did the phishing email arrive and the attachment open?</li><li>Which systems are still down, and what is the realistic resort time?</li><li>Has the attacker made a ransom demand or a threat to publish data?</li></ul><p><em>Assumptions and Limitations</em></p><ul><li>“Monday” is taken as the most recent Monday, October 19, 2026.</li><li>Attack steps beyond the stated IOCs (lateral movement, exfiltration) are inferences.</li><li>Containment actions are recommendations; completion status is not in the source information. Legal references are indicative and need confirmation from counsel.</li></ul><p><em>Next Steps (tick oﬀ as completed)</em></p><ul><li>Executive team to approve external incident response and legal engagement.</li><li>Security team to complete log analysis and update this report with confirmed facts.</li><li>Legal to confirm notification duties and deadlines in Australia and New Zealand.</li><li>Schedule a post-incident review since systems are restored.</li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=ff74acd2c572" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[SOHO Network Lab (Cisco CLI)]]></title>
            <link>https://medium.com/@willedwardsjr79/soho-network-lab-cisco-cli-d629dc30b8c3?source=rss-f4eceb10dfa9------2</link>
            <guid isPermaLink="false">https://medium.com/p/d629dc30b8c3</guid>
            <category><![CDATA[ccna-training]]></category>
            <category><![CDATA[cisco-packet-tracer]]></category>
            <category><![CDATA[cisco]]></category>
            <dc:creator><![CDATA[Will Edwards]]></dc:creator>
            <pubDate>Fri, 09 Oct 2026 23:59:38 GMT</pubDate>
            <atom:updated>2026-10-09T23:59:38.991Z</atom:updated>
            <content:encoded><![CDATA[<p>Overview:<br>This lab builds a small-office network with staff, guest, and management VLANs, NAT to a simulated internet service provider., a guest-isolation access control list (ACL), a network printer, and a wireless access point (WAP). Plan on about two hours in Packet Tracer, GNS3, CML, or on real 2900/2960-class gear.</p><p>Devices:<br>One ISP router, one SOHO router (1941, 2911, 0r 4321), one 2960 switch, two or three PCs, one printer, and one access point (two in Packet Tracer).</p><p>Objectives:<br> 1. Build the topology and set base configuration<br> 2. Creat VLANs and a trunk<br> 3. Configure router-on-a-stick inter-VLAN routing<br> 4. Configure DHCP<br> 5. Configure NAT overload and a default route<br> 6. Isolate guest with an ACL<br> 7. Harden the switch and router<br> 8. Verify and troubleshoot<br> 9. Add a network printer<br>10. Add a WAP with staff and guest SSIDs</p><p>Addressing Plan<br>Three VLANS share one router interface, and every host addres sits in a /24 CIDR except the point-to-point WAN link.<br>Staff Network: VLAN 10; Subnet 192.168.10.0/24; Gateway .1<br>Guest Network: VLAN 20; Subnet 192.168.20.0/24; Gateway .1<br>Management Network: VLAN 99; Subnet 192.168.99.0/24 Gateway .1 (switch is .2, AP is .3)<br>WAN Link Network: VLAN n/a; Subnet 203.0.113.0/30; Gateway ISP .1, SOHO-R .2</p><p>Device<br>Printer VLAN 10; IP 192.168.10.10; Static, inside the DHCP-excluded range<br>AP Management VLAN 99 (mgmt); IP 192.168.99.3; Static<br>Wireless staff clients VLAN 10; IP 192.168.10.21 and up; DHCP<br>Wireless guest clients VLAN 20; IP 192.168.20.21 and up; DHCP</p><p>Basic Commands for all devices at start up<br>Once each device is powers on, if it ask you if you would like to enter intial configuration dialog, press n. At the &gt; prompt type en to switch to the # prompt, then type “conf t” and press enter to access the config terminal prompt.</p><p><strong>Steps 1 to 3: Build the core</strong><br>These steps bring up the ISP stand-in, the switch VLANs and trunk, and inter-VLAN routing on the SOHO router.</p><p><strong>Step 1: ISP side</strong><br>Loopback0 stands for the “internet.” The ISP needs no return route because the SOHO router will NAT everything to 203.0.113.2</p><p>Router(config)# hostname ISP<br>ISP(config)# interface g0/0<br>ISP(config)# ip address 203.0.113.1 255.255.255.252<br>ISP(config)# no shutdown<br>ISP(config)# interface loopback0<br>ISP(config)# ip address 8.8.8.8 255.255.255.255</p><p><strong>Step 2: Switch VLANs and trunk</strong><br>Use the switch’s actual uplink name (Fa0/2 on many 2960s). On a 3560 or 3750, add switchport trunk encapsulation dot1q first.</p><p>Switch(config)# hostname SW1<br>SW1(config)# vlan 10<br>SW1(config-vlan)# name STAFF<br>SW1(config)# Alan 20<br>SW1(config-vlan)# name GUEST<br>SW1(config)# vlan 99<br>SW1(config-vlan)# name MGMT<br>SW1(config)# vlan 999<br>SW1(config-vlan)# name PARKING</p><p>SW1(config)# interface g0/1<br>SW1(config-if)# description Trunk to SOHO-R<br>SW1(config-if)# switchport mode trunk<br>SW1(config-if)# switchport trunk native vlan 99<br>SW1(config-if)# switchport trunk aloowed clan 10, 20, 99</p><p>SW1(config)# interface range f0/2–10<br>SW1(config-if-range)# switchport mode access<br>SW1(config-if-range)# switchport access vlan 10<br>SW1(config)# interface range f0/11–20<br>SW1(config-if-range)# switchport mode access<br>SW1(config-if-range)# switchport access vlan 20</p><p>SW1(config)# interface vlan 99<br>SW1(config-if)# ip address 192.168.99.2 255.255.255.0<br>SW1(config-if)# not shutdown<br>SW1(config)# ip default gateway 192.168.99.1</p><p><strong>Step 3: Router-on-a-stick</strong><br>Router(config)# hostname SOHO-R<br>SOHO-R(config)# interface g0/1<br>SOHO-R(config-if)# no shutdown<br>SOHO-R(config)# interface g0/1.10<br>SOHO-R(config-subif)# encapsulation dot1Q 10<br>SOHO-R(config-subif)# ip address 192.168.10.1 255.255.255.0<br>SOHO-R(config-subif)# ip nat inside<br>SOHO-R(config)# interface g0/1.20<br>SOHO-R(config-subif)# encapsulation dot1Q 20<br>SOHO-R(config-subif)# ip address 192.168.20.1 255.255.255.0<br>SOHO-R(config-subif)# ip nat inside<br>SOHO-R(config)# interface g0/1.99<br>SOHO-R(config-subif)# encapsulation dot1Q 99 native<br>SOHO-R(config-subif)# ip address 192.168.99.1 255.255.255.0</p><p><strong>Checkpoint</strong>: show ip interface brief on SOHO-R shows all subinterfaces up/up, and SW1 can ping 192.168.99.1.</p><p><strong>Steps 4 to 6: Services and Isolation<br></strong>These steps hand out addresses, give the LANs internet access through NAT, and keepguests away from staff and management.</p><p><strong>Step 4: DHCP</strong><br>SOHO-R(config)# ip dhcp excluded-address 192.168.10.1 192.168.10.20<br>SOHO-R(config)# ip dhcp excluded-address 192.168.20.1 192.168.20.20<br>SOHO-R(config)# ip dhcp pool STAFF<br>SOHO-R(dhcp-config)# network 192.168.10.0 255.255.255.0<br>SOHO-R(dhcp-config)# default-router 192.168.10.1<br>SOHO-R(dhcp-config)# dns-server 8.8.8.8<br>SOHO-R(config)# ip dhcp pool GUEST<br>SOHO-R(dhcp-config)# network 192.168.20.0 255.255.255.0<br>SOHO-R(dhcp-config)# default-router 192.168.20.1<br>SOHO-R(dhcp-config)# dns-server 8.8.8.8<br>Set PC1 and PC2 to DHCP. <br><strong>Checkpoint:</strong> PC1 gets 192.168.10.21 or higher, and PC2 gets 192.168.20.21 or higher.</p><p><strong>Step 5: WAN, default route, and NAT<br></strong>SOHO-R(config)# interface g0/0<br>SOHO-R(config-if)# description WAN to ISP<br>SOHO-R(config-if)# ip address 203.0.113.2 255.255.255.252<br>SOHO-R(config-if)# ip nat outside<br>SOHO-R(config-if)# no shutdown<br>SOHO-R(config)# ip route 0.0.0.0 0.0.0.0 203.0.113.1<br>SOHO-R(config)# access-list 1 permit 192.168.10.0 0.0.0.255<br>SOHO-R(config)# access-list 1 permit 192.168.20.0 0.0.0.255<br>SOHO-R(config)# ip nat inside source list 1 interface g0/0 overload<br><strong>Checkpoint:</strong> PC1 pings 8.8.8.8. Then run show ip nat translations.</p><p><strong>Step 6: Guest isolation ACL<br></strong>SOHO-R(config)# ip access-list extended GUEST-IN<br>SOHO-R(config-ext-nacl)# deny ip 192.168.20.0 0.0.0.255 192.168.10.0 0.0.0.255<br>SOHO-R(config-ext-nacl)# deny ip 192.168.20.0 0.0.0.255 192.168.99.0 0.0.0.255<br>SOHO-R(config-ext-nacl)# permit ip any any<br>SOHO-R(config)# interface g0/1.20<br>SOHO-R(config-subif)# ip access-group GUEST-IN in<br><strong>Checkpoint:</strong> PC2 can ping 8.8.8.8 but not PC1 or 192.168.99.2. PC1 can still reach everything.</p><p><strong>Step 7: Hardening<br></strong>Harden both devices with SSH-only access and local accounts, then add Layer 2 protections on the switch and a management-VLAN restriction on the router.</p><p><em>Both devices</em><br>(config)# enable secret Str0ngEnable!<br>(config)# service password-encryption<br>(config)# no ip domain-lookup<br>(config)# banner motd # Authorized access only #<br>(config)# ip domain-name soho.local<br>(config)# username admin secret Str0ngAdmin!<br>(config)# crypto key generate rsa modulus 2048<br>(config)# ip ssh version 2<br>(config)# line vty 0 4<br>(config-line)# transport input ssh<br>(config-line)# login local<br>(config-line)# exec-timeout 5 0<br>(config)# line console 0<br>(config-line)# logging synchronous<br>(config-line)# login local</p><p><em>Switch only<br></em>SW1(config)# interface range f0/2–20<br>SW1(config-if-range)# spanning-tree portfast<br>SW1(config-if-range)# spanning-tree bpduguard enable<br>SW1(config-if-range)# switchport port-security<br>SW1(config-if-range)# switchport port-security maximum 2<br>SW1(config-if-range)# switchport port-security mac-address sticky<br>SW1(config-if-range)# switchport port-security violation restrict<br>SW1(config)# interface range f0/21–24, g0/2<br>SW1(config-if-range)# switchport mode access<br>SW1(config-if-range)# switchport access vlan 999<br>SW1(config-if-range)# shutdown</p><p>Adjust the interface ranges to match your switch model. In Packet Tracer, some of these commands (BPDU guard, sticky) may be limited.</p><p><em>Router only:</em> restrict SSH to the management VLAN.<br>SOHO-R(config)# access-list 10 permit 192.168.99.0 0.0.0.255<br>SOHO-R(config)# line vty 0 4<br>SOHO-R(config-line)# access-class 10 in</p><p>Management then happens from a device in VLAN 99. Test from a host there before you apply it, or you may lock yourself out.</p><p>Save the config with copy running-config startup-config on every device.</p><p><strong>Step 8: Verify the base lab</strong></p><p>Run each test before moving on to the printer and wireless steps.</p><p>Test: show vlan brief (SW1)<br>Expected: VLANs 10, 20, 99, and 999 with the right ports</p><p>Test: show interfaces trunk (SW1)<br>Expected: g0/1 trunking, native 99, allowed 10,20,99</p><p>Test: show ip interface brief (SOHO-R)<br>Expected: All interfaces up/up</p><p>Test: show ip route <br>Expected: S* default route via 203.0.113.1</p><p>Test: show ip dhcp binding<br>Expected: Leases for PC1 and PC2</p><p>Test: show ip nat translations and statistics<br>Expected: Active overload entries, hits increasing</p><p>Test: Show access-lists<br>Expected: Deny counters rise when PC2 pings PC1</p><p><strong>Step 9: Add the printer</strong></p><p>The printer sits on a staff port and gets a fixed address, so drivers and print queues don’t break when a lease changes. Fa0/5 is already in the Fa0/2–10 range, so its VLAN is set.</p><p>Printer settings (Packet Tracer: Printer-PT, Config, FastEthernet):<br>IP address: 192.168.10.10<br>Subnet mask: 255.255.255.0<br>Gateway: 192.168.10.1</p><p>The address falls inside ip dhcp excluded-address 192.168.10.1 192.168.10.20, so</p><p>DHCP will never hand it to another device.</p><p>Alternative: DHCP reservation. Use this on real gear if the printer insists on DHCP. Get the MAC from show mac address-table interface f0/5.</p><p>SOHO-R(config)# ip dhcp pool PRINTER<br>SOHO-R(dhcp-config)# host 192.168.10.10 255.255.255.0<br>SOHO-R(dhcp-config)# client-identifier 01aa.bbcc.ddee.ff<br>SOHO-R(dhcp-config)# default-router 192.168.10.1</p><p>For a client-identifier, prefix the MAC with 01. On routers that accept it, hardware-address aabb.ccdd.eeff ethernet works instead.</p><p><strong>Checkpoint:</strong> PC1 pings 192.168.10.10. PC2 (guest) cannot, and the GUEST-IN deny counter increases.</p><p><strong>Steps 10 and 11: Wireless AP and hardening</strong></p><p>The AP port must be a trunk (or one access port per AP in Packet Tracer), and it must not inherit port security or BPDU guard. Many wireless client MACs appear behind one port, and port security would shut it down. Fa0/21 was also in the parking range from Step 7, so pull it out of that range.</p><p>SW1(config)# interface f0/21<br>SW1(config-if)# description Trunk to AP<br>SW1(config-if)# no shutdown<br>SW1(config-if)# switchport mode trunk<br>SW1(config-if)# switchport trunk native vlan 99<br>SW1(config-if)# switchport trunk allowed vlan 10,20,99<br>SW1(config-if)# switchport nonegotiate</p><p>On a 3560 or 3750, add switchport trunk encapsulation dot1q first. On a 2960, Fa0/21-24 were set to access VLAN 999 and shut down in Step 7. The commands above override that for Fa0/21 only.</p><p><em>Option A: Packet Tracer (simplest)</em></p><p>Packet Tracer’s Access Point-PT can’t tag VLANs per SSID, so use two APs, one per VLAN, on access ports instead of a trunk. If you do this, skip the Fa0/21 trunk above.</p><p>SW1(config)# interface f0/3<br>SW1(config-if)# description AP-STAFF<br>SW1(config-if)# switchport access vlan 10<br>SW1(config)# interface f0/12<br>SW1(config-if)# description AP-GUEST<br>SW1(config-if)# switchport access vlan 20</p><p>On each AP (Config, Port 1 / Wireless):</p><p>Setting AP-STAFF AP-GUEST<br>SSID STAFF GUEST<br>Security WPA2-PSK WPA2-PSK<br>Passphrase StaffWiFi!2026 GuestWiFi!2026</p><p>Add two laptops with wireless NICs (swap the PT-LAPTOP’s Ethernet module for WPC300N), set them to DHCP, and join each to its SSID.</p><p><em>Option B: One autonomous AP, two SSIDs (real gear or CML)</em></p><p>This is closer to a real SOHO deployment. Syntax varies by AP model and IOS version, and the GUI (Express Setup, then Security, then SSID Manager) generates the bridging config for you. The core pieces on an autonomous Cisco IOS AP:</p><p>AP(config)# hostname AP1<br>AP1(config)# dot11 ssid STAFF<br>AP1(config-ssid)# vlan 10<br>AP1(config-ssid)# authentication open<br>AP1(config-ssid)# authentication key-management wpa version 2<br>AP1(config-ssid)# wpa-psk ascii StaffWiFi!2026</p><p>AP1(config)# dot11 ssid GUEST<br>AP1(config-ssid)# vlan 20<br>AP1(config-ssid)# authentication open<br>AP1(config-ssid)# authentication key-management wpa version 2<br>AP1(config-ssid)# wpa-psk ascii GuestWiFi!2026</p><p>AP1(config)# interface Dot11Radio0<br>AP1(config-if)# encryption vlan 10 mode ciphers aes-ccm<br>AP1(config-if)# encryption vlan 20 mode ciphers aes-ccm<br>AP1(config-if)# ssid STAFF<br>AP1(config-if)# ssid GUEST<br>AP1(config-if)# mbssid<br>AP1(config-if)# no shutdown</p><p>Then create matching subinterfaces (.10, .20, .99) on both Dot11Radio0 and GigabitEthernet0, each with encapsulation dot1Q &lt;vlan&gt; (add native on .99) and a bridge-group &lt;vlan&gt;. Give the management bridge an address:</p><p>AP1(config)# interface BVI1<br>AP1(config-if)# ip address 192.168.99.3 255.255.255.0<br>AP1(config)# ip default-gateway 192.168.99.1</p><p>Lightweight APs under a WLC work differently, since the WLC handles the SSID-to-VLAN mapping. That is a separate lab.</p><p><strong>Step 11: Wireless hardening</strong></p><p>Use WPA2-PSK at minimum, and WPA3 where supported. Never leave the guest SSID open.<br>Use a different passphrase per SSID.<br>Enable client isolation on the guest SSID so guests can’t reach each other (“Public Secure Packet Forwarding” or “Client Isolation” in the GUI).<br>Disable WPS and set a strong admin password on the AP.</p><p>Don’t map the management VLAN to any SSID. The AP’s own management traffic uses native VLAN 99.</p><p><strong>Step 12: Verify the printer and wireless</strong></p><p>Option A rows apply to the two-AP Packet Tracer build; Option B rows apply to the single trunked AP.</p><p>Test: show interfaces trunk (SW1), Option B<br>Expected: Fa0/21 trunking, native 99, allowed 10,20,99</p><p>Test: show vlan brief, Option A <br>Expected: Fa0/3 in VLAN 10, Fa0/12 in VLAN 20</p><p>Test: show mac address-table<br>Expected: Printer MAC on Fa0/5 in VLAN 10, AP MAC on its port</p><p>Test: Staff laptop joins STAFF<br>Expected: Gets 192.168.10.21 or higher and pings 8.8.8.8</p><p>Test: Guest laptop joins GUEST<br>Expected: Gets 192.168.20.21 or higher and pings 8.8.8.8</p><p>Test: Staff laptop to printer (192.168.10.10)<br>Expected: Success</p><p>Test: Guest laptop to printer or PC1<br>Expected: Fails; ACL counters increase</p><p>Test: show ip dhcp binding<br>Expected: Wireless clients appear in the correct pools</p><p>Test: Ping 192.168.99.3 from VLAN 99<br>Expected: AP management is reachable</p><p><strong>Troubleshooting scenarios</strong></p><p>Break the working lab on purpose, find the fault with show commands, then fix it.</p><p><em>Base lab</em></p><p>1. No DHCP on one VLAN: remove the encapsulation dot1Q 20 line. Which symptom appears first?<br>2. Trunk mismatch: change the native VLAN on only one side and watch for CDP and STP native VLAN mismatch messages.<br>3. No internet but local works: remove ip nat outside from g0/0, then use show ip nat statistics to find it.<br>4. Guests can reach staff: apply the ACL out instead of in, or put it on the wrong subinterface.<br>5. Port-security shutdown: plug a second device into a staff port with maximum 1 and violation shutdown, then recover with shutdown and no shutdown.</p><p><em>Printer and wireless</em></p><p>1. Wireless client gets a 169.254.x.x address: remove VLAN 20 from the trunk’s allowed list (Option B), or move Fa0/12 to the wrong VLAN (Option A).<br>2. AP port goes err-disabled: apply port security with maximum 1 to the AP port and connect two wireless clients. Check show interfaces status err-disabled, then recover with shutdown and no shutdown.<br>3. Printer is unreachable from staff: change the printer’s gateway or mask and see which hosts can still reach it. Same-VLAN hosts work; routed ones fail.<br>4. Guest can print: remove the deny … 192.168.10.0 line from GUEST-IN and confirm the printer becomes reachable, then restore it.<br>5. Duplicate IP: set the printer to 192.168.10.25 (outside the excluded range) and watch DHCP hand that address to a PC.</p><p><strong>Stretch goals<br>-</strong>Replace router-on-a-stick with a Layer 3 switch (SVIs plus ip routing).<br>-Add static NAT for a server or the printer.<br>-Allow only TCP 9100 and 631 from staff to the printer, and block everything else to it with an ACL.<br>-Put the printer on its own VLAN (for example VLAN 30) with a printer-only ACL.<br>-Use WPA2-Enterprise with a RADIUS server for the staff SSID.<br>-Add a second AP and tune channels (1, 6, 11) to avoid overlap.<br>-Add syslog and NTP, then SNMPv3 read-only monitoring.<br>-Add IPv6 dual-stack on the staff VLAN.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=d629dc30b8c3" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>