Blog - Edgeium

Can Cisco Access Switches Stay Secure and Compliant Without New IOS Updates

Written by Eric Sommers | 01.15.2026

Are software updates required to maintain compliance with SOC 2, ISO 27001, or PCI DSS?

The short answer:   No, a properly hardened Cisco access-layer switch can remain secure and compliant without requiring a newer IOS image.  The compliance requirement is effective risk management: identifying known vulnerabilities, determining whether they are applicable and reachable, reducing or removing the exposure, and maintaining evidence of the controls in place.   For stable access-layer switches, that can be achieved through deliberate LAN configuration hardening: least privilege, strong authentication, SSH-only management, management ACLs, VLAN segmentation, disabled unused services, centralized logging, configuration retention, control-plane protection, and documented change control. 

 

Compliance requires vulnerability management, controls, and evidence

The compliance requirement is not continuous access to every future IOS image. The compliance requirement is documented vulnerability management, effective controls, and evidence. The question is not whether a newer IOS image exists. The question is whether the switch has an applicable, reachable, unmitigated vulnerability that creates real exposure in its deployed role.

A Cisco advisory may list an IOS or IOS XE release as affected, but that does not automatically mean every access switch running that release is exposed. If the vulnerable service is disabled, unreachable, restricted by management ACLs, isolated to a management VLAN, or mitigated through configuration and architecture, the practical attack path may already be removed.

Access-layer switches are different because access-layer exposure is different

Where the device sits in the network matters. Core, security, data center, wireless controller, and internet-facing platforms may justify a much closer software-update path.  Access-layer switches are different. They usually sit at the edge of the LAN, provide endpoint connectivity, and operate behind multiple layers of network security, segmentation, monitoring, and policy enforcement.

Access-layer environments also create a different operational reality. Access switches often represent the largest portion of a network estate, sometimes 60–70% of all network assets. This article is focused on those last device connections: closet switches, access-layer switches, and the infrastructure providing network access to endpoints, phones, cameras, access points, printers, and other edge-connected devices.

The FUD objection treats "affected" as "exposed"

We continue to see software updates create unexpected network issues. Software updates are important for certain but not all platforms, and they should be treated like any other production change: tested, justified, scheduled, and tied to a real requirement.

This is where the FUD Monster usually shows up. When discussing secondary-market hardware or alternatives to OEM maintenance, one of the most common objections is the “need to have access to IOS updates.” The implication is simple: if you do not have access to the newest IOS image, the device cannot be secure or compliant.  But for hardened access-layer switches, that conclusion is inaccurate.

The FUD objection treats “affected” as “exposed.” But affected is only the first question. The better analysis is whether the vulnerability is applicable, reachable, unmitigated, and material to the switch’s deployed role.

Continuous image updates is not a practical access-layer control

In large access-layer environments, maintaining every switch on the newest available image is often operationally unrealistic. 

From 2020-2025, 148 "High" security vulnerabilities were issued for Catalyst IOS-XE builds.  To comply with the notion that someone out there is current, that would have required a NetOps team to have made 29.6 (unexpected or unplanned) image changes per year.  That's 2.5 image changes per month.  Now multiply this by the number of switches you have in your network.  And don't forget at least 2 weeks per new image to work through change management safeguards.  

The reason why most organizations have not historically staffed dedicated teams to keep every access switch on the latest image is that the operational return hasn't justified the cost - especially when the platform is stable, hardened, and not exposed to the vulnerable service path.

Hardening reduces or removes attack paths

For the access layer, the practical security model should start with hardening:

  • Disable unused services

  • Disable HTTP/HTTPS management when not required

  • Disable Smart Install when not required

  • Use SSH-only management

  • Restrict management access with ACLs

  • Isolate management VLANs

  • Enforce AAA and least privilege

  • Segment user, voice, wireless, guest, and management traffic

  • Use port security, DHCP snooping, dynamic ARP inspection, BPDU Guard, and storm control

  • Send logs to centralized syslog with reliable timestamps

  • Retain configuration history and change evidence 

 When these controls are in place, the access switch should be evaluated by actual exposure, not IOS version alone. Security comes from reducing attack surface, controlling management access, documenting controls, and addressing real exposure.

 If a properly hardened access-layer switch is stable, controlled, monitored, documented, and free of applicable, reachable, unmitigated vulnerabilities, it can remain secure and compliant without requiring a newer IOS image. 

Web UI and Smart Install show why exposure matters

Example 1: IOS XE Web UI privilege escalation (CVE-2023-20198), CVSS 10.0, the maximum score possible

CVE-2023-20198 was a critical IOS XE Web UI privilege-escalation vulnerability with a CVSS score of 10.0. It received significant attention because it allowed an unauthenticated remote attacker to create a high-privilege local account on an affected device.

In October 2023 this zero-day let an unauthenticated, remote attacker create a full privilege-level-15 account and take over the device. It was mass-exploited and tens of thousands of devices were impacted within days, but here is the detail that matters: the vulnerability only affects devices with the HTTP or HTTPS server feature enabled, and the mitigation, per Cisco and CISA, was to disable it with no ip http server and no ip http secure-server, or restricting access with HTTP ACLs when the feature could not be fully disabled

Look back at Section 1 of the hardened configuration above. Those two commands are already there. A switch hardened to this template  would not have exposed that Web UI attack path, because HTTP and HTTPS management were already disabled. The follow-on command-injection flaw in the same chain (CVE-2023-20273, CVSS 7.2) depended on the same Web UI being reachable.

That distinction matters. A hardened access-layer switch with HTTP and HTTPS management disabled, management access restricted, and no reachable Web UI attack path should not be treated the same as an exposed device simply because both devices appear in the same software advisory.

Example 2: Cisco Smart Install remote code flaw (CVE-2018-0171), CVSS 9.8
CVE-2018-0171 was a high-severity Smart Install remote code execution vulnerability. The risk depended on the Smart Install client feature being enabled and reachable on TCP port 4786.  This allowed unauthenticated remote code execution by sending a crafted Smart Install message to TCP port 4786, and it is still being actively exploited by state-sponsored actors years later.

Again, the exposure was conditional.  It required the Smart Install client to be enabled and reachable. A hardened access switch disables it (no vstack) and, with a management ACL restricting which hosts can even reach the control plane, removes the attack surface entirely.  

"No workaround" still requires analysis, not  an automatic update

Some Cisco Catalyst advisories state that no workaround is available while still describing exposure conditions, recovery steps, or containment actions. That usually means Cisco is not providing a vendor-validated configuration workaround that fully fixes the software defect in the running image. It does not automatically mean every device running that image is exposed, exploitable, or non-compliant.

Depending on the advisory, remediation may include:

  • Confirming whether the vulnerable feature is enabled

  • Disabling the vulnerable or unused service

  • Restricting reachability with management ACLs, VLAN ACLs, MAC ACLs, or control-plane policy

  • Isolating the management plane

  • Removing or blocking the source of malicious traffic

  • Changing the configuration so the affected condition no longer exists

  • Moving the device to a validated non-affected image, which may be newer, older, or simply different depending on the advisory

  • Documenting temporary risk treatment until final remediation is complete

     

Affected is not the same as exposed. Exposed is not the same as unmitigated. Unmitigated is where security and compliance risk begins

The config and compliance table provide the evidence

The framework above only works if it can be documented. That is why hardening should be paired with configuration evidence, vulnerability review, and a configuration attestation memo.

A hardened LAN configuration is a set of deliberate security controls applied to access-layer switches to reduce attack surface and prevent common LAN-based risks such as rogue device insertion, VLAN hopping, spoofing, unauthorized management access, and control-plane exposure.

For Cisco access-layer switches, those controls may include least privilege, strong authentication, SSH-only management, management ACLs, VLAN segmentation, disabled unused services, centralized logging, configuration retention, control-plane protection, and documented change control.

The examples below show how those controls can be mapped to SOC 2 Trust Services Criteria, ISO 27001 control expectations, and PCI DSS security objectives without treating continuous access to every future IOS image as the compliance control.

Compliance Summary Table

Framework control Requirement Addressed by configuration / evidence
SOC 2 CC6.1 Logical access controls and secure configuration System identification, baseline hardening, SSH-only management, management ACLs, disabled unused services
SOC 2 CC6.2 User authentication and authorization AAA, local authentication, privilege levels, least-privilege administrative access
SOC 2 CC6.3 Protection against unauthorized access and brute-force attempts Login block-for, login delay, strong secrets, restricted VTY access
SOC 2 CC6.6 Network segmentation and logical separation User, voice, WiFi, guest, management, and inactive-port VLANs; VLAN 1 shutdown; trunk controls
SOC 2 CC6.7 Secure administrative access SSH v2, RSA keys, transport input SSH, VTY ACL, management VLAN isolation
SOC 2 CC7.1 Preventive security controls No IP source-route, no HTTP/HTTPS server when not required, no Smart Install, BPDU Guard, storm control, DHCP snooping, dynamic ARP inspection
SOC 2 CC7.2 Detection of anomalies and security events Syslog host, logging trap, authentication logs, control-plane events, configuration-change logging
SOC 2 CC7.3 Security event evaluation and response CVE review, affected-version analysis, feature exposure review, reachability assessment, documented remediation decision
SOC 2 CC7.4 Logging consistency for investigation Timestamps with milliseconds, logging buffered, centralized syslog, retained event history
SOC 2 CC8.1 Change management controls Archive log config, notify syslog, saved configuration, change records, configuration attestation memo
SOC 2 CC9.2 Risk mitigation Documented risk treatment for applicable vulnerabilities, compensating controls, isolation, replacement path, or validated non-affected image
ISO 27001:2022 A.5.9 Inventory of information and associated assets Switch inventory, serial number, platform, IOS version, site, role, management IP, support path
ISO 27001:2022 A.8.8 Management of technical vulnerabilities Vulnerability review, applicability assessment, reachable-service review, mitigation evidence, remediation tracking
ISO 27001:2022 A.8.9 Configuration management Approved hardened baseline, configuration archive, standard templates, change history
ISO 27001:2022 A.8.15 Logging Syslog, timestamps, logging buffer, event forwarding, log retention
ISO 27001:2022 A.8.16 Monitoring activities Monitoring of switch availability, authentication attempts, configuration changes, and control-plane events
ISO 27001:2022 A.8.20 Network security Management ACLs, secure protocols, control-plane protection, disabled unused services
ISO 27001:2022 A.8.21 Security of network services SSH-only management, restricted SNMP, disabled HTTP/HTTPS, authenticated NTP, service minimization
ISO 27001:2022 A.8.22 Segregation of networks VLAN segmentation, management VLAN isolation, inactive-port VLAN, trunk restrictions
PCI DSS 4.0.1 Req. 1.2 / 1.3 Network security controls are configured and traffic is restricted VLAN isolation, trunk protection, management ACLs, CDE isolation, documented allowed paths
PCI DSS 4.0.1 Req. 2.2 Hardened system configuration Disabled unused services, strong authentication, secure management protocols, documented baseline configuration
PCI DSS 4.0.1 Req. 6.3 Vulnerabilities are identified, risk-ranked, and addressed CVE review, applicability assessment, exposure determination, mitigation evidence, remediation plan
PCI DSS 4.0.1 Req. 10 Log and monitor access to system components Syslog, timestamps, authentication logs, configuration-change logging, log review evidence
PCI DSS 4.0.1 Req. 11.3 Internal vulnerability scanning, where in scope Scan results, exception documentation, mitigation evidence, remediation tracking

Configuration Example

Implementation note

Validate before deploying

The following configuration is an example control set, not a universal paste-in template. Commands should be validated against the platform, IOS version, site design, voice/WiFi requirements, multicast requirements, monitoring tools, and compliance scope before deployment.

Get config hardening help →

The following example shows common hardening controls for Cisco access-layer switches. It is intended to demonstrate how configuration evidence can support documented vulnerability management, effective controls, and audit readiness.

1. System identification, configuration integrity, and baseline hardening

hostname SW-3850-Hardened-Config
service password-encryption
no service pad
no service tcp-small-servers
no service udp-small-servers
no service config
no ip source-route
no ip finger
no ip rcmd rcp-enable
no ip rcmd rsh-enable
no ip domain-lookup
no ip gratuitous-arps
no ip http server
no ip http secure-server

! Optional: disable only if not required for the site design
! no cdp run
! no lldp run

! Optional: use only if the switch is not providing DHCP services
! no service dhcp

2. Time synchronization and logging integrity

service timestamps debug datetime msec localtime
service timestamps log datetime msec localtime

ntp authenticate
ntp authentication-key 1 md5 <KEY>
ntp trusted-key 1
ntp server <IP ADDRESS> key 1

3. Authentication, authorization, and account security

aaa new-model
aaa authentication login default local
aaa authorization exec default local
aaa authentication login CONSOLE local
aaa authentication login VTY local
aaa authorization exec VTY local
aaa session-id common

username admin privilege 15 secret <PASSWORD>
enable secret <PASSWORD>

login block-for 120 attempts 3 within 60
login delay 5

4. Secure management access with SSH and ACLs

ip domain-name edgeium.local
crypto key generate rsa modulus 4096

ip ssh version 2
ip ssh time-out 30
ip ssh authentication-retries 2

ip access-list standard MGMT-ACL
 permit 10.10.10.0 0.0.0.255
 deny any log

5. VTY enforcement for SSH-only management

line vty 0 4
 access-class MGMT-ACL in
 transport input ssh
 exec-timeout 10 0
 logging synchronous

line vty 5 15
 access-class MGMT-ACL in
 transport input ssh
 exec-timeout 10 0
 logging synchronous

6. SNMP restriction or removal

no snmp-server community public
no snmp-server community private

! If SNMP is required, use SNMPv3 and restrict polling sources:
snmp-server group EDGEIUM-MON v3 priv access MGMT-ACL
snmp-server user <USER> EDGEIUM-MON v3 auth sha <AUTH-KEY> priv aes 256 <PRIV-KEY>

7. Centralized logging and audit trails

logging buffered 64000 warnings
logging host <IP ADDRESS>
logging trap warnings
logging console critical
logging monitor warnings

archive
 log config
  logging enable
  notify syslog

8. Network segmentation and management isolation

vlan 10
 name USERS

vlan 20
 name VOICE

vlan 30
 name WIFI

vlan 40
 name GUEST

vlan 99
 name MGMT

vlan 998
 name UNUSED_NATIVE_VLAN

vlan 999
 name INACTIVE_PORTS

interface vlan 1
 shutdown

9. Access-port protections

interface range <ACCESS-PORT-RANGE>
 description USER ACCESS PORTS
 switchport mode access
 switchport access vlan <USER-VLAN>
 switchport voice vlan <VOICE-VLAN>
 switchport port-security
 switchport port-security maximum 2
 switchport port-security violation restrict
 switchport port-security mac-address sticky
 spanning-tree portfast
 spanning-tree bpduguard enable
 storm-control broadcast level 1.00
 storm-control multicast level 1.00

10. DHCP Snooping and Dynamic ARP Inspection

ip dhcp snooping
ip dhcp snooping vlan <USER-VLANS>
ip arp inspection vlan <USER-VLANS>

interface <UPLINK-INTERFACE>
 description TRUSTED UPLINK TO DISTRIBUTION
 ip dhcp snooping trust
 ip arp inspection trust

interface range <ACCESS-PORT-RANGE>
 description UNTRUSTED ACCESS PORTS
 no ip dhcp snooping trust
 no ip arp inspection trust

11. Trunk hardening

interface <TRUNK-INTERFACE>
 description UPLINK TO DISTRIBUTION
 switchport mode trunk
 switchport trunk native vlan 998
 switchport trunk allowed vlan <APPROVED-VLANS>
 switchport nonegotiate

12. Control-plane protection placeholder

ip access-list extended CP-MGMT
 permit tcp 10.10.10.0 0.0.0.255 any eq 22
 deny ip any any log

class-map match-any MGMT-TRAFFIC
 match access-group name CP-MGMT

policy-map CONTROL-PLANE-POLICY
 class MGMT-TRAFFIC
  police <RATE>
 class class-default
  police <RATE>

control-plane
 service-policy input CONTROL-PLANE-POLICY

13. Configuration retention

write memory

14. Evidence collection for configuration attestation

show version
show inventory
show running-config
show ip interface brief
show ip ssh
show access-lists
show running-config | section line vty
show running-config | include ip http|vstack|snmp-server|service
show logging
show archive
show ip dhcp snooping
show ip arp inspection
show spanning-tree summary

The configuration example above is only partially complete. A final hardened template should be tailored to the platform, IOS version, site design, management model, monitoring tools, endpoint requirements, and compliance scope.

For audit evidence, the configuration should be paired with a configuration attestation memo that documents the device role, current IOS image, hardening controls, vulnerability review, applicable exceptions, compensating controls, and remediation decisions.

That evidence package supports the article’s core point: the compliance requirement is not continuous access to every future IOS image. The compliance requirement is documented vulnerability management, effective controls, and evidence.

Thank you for reading!

One specific next step

Planning an access-layer refresh right now?

Before you replace your access switches, get a second opinion. Edgeium reviews refresh timing, hardware alternatives, support exposure, and licensing — to show what actually needs replacing, and where it does not.

No obligation · Your current VAR stays your VAR

Frequently Asked Questions

Do Cisco access-layer switches need IOS updates to meet PCI DSS requirements? 

No, not by default. PCI DSS requires organizations to identify, risk-rank, and address applicable vulnerabilities. A newer IOS image is not automatically required if the vulnerability is not applicable, not reachable, or has been mitigated through hardening, segmentation, access control, disabled services, compensating controls, or documented risk treatment. If a vulnerability remains applicable, reachable, and unmitigated, the affected asset should be patched, isolated, moved to a validated non-affected image, replaced, or otherwise remediated. 

What is LAN hardening and how does it help with compliance? 

LAN hardening is the practice of applying deliberate security configurations to network switches to reduce attack surface and prevent common LAN-based risks such as rogue device insertion, VLAN hopping, spoofing, unauthorized management access, and control-plane exposure. A hardened access-layer configuration may include least privilege, AAA, SSH-only management, management ACLs, VLAN segmentation, inactive port isolation, port security, DHCP snooping, dynamic ARP inspection, BPDU Guard, storm control, centralized syslog, reliable timestamps, and configuration-change evidence. 

Why is IOS version not the compliance control? 

IOS version alone is not the compliance control because a listed vulnerability does not automatically create exposure. The compliance question is whether the vulnerability is applicable to the platform, reachable in the deployed environment, unmitigated by configuration or architecture, and material to the device’s role. The control is documented vulnerability management, effective hardening, risk treatment, and evidence. 

What does affected but not exposed mean for Cisco switch vulnerabilities?

Affected means a platform or software release appears in a vendor advisory. Exposed means the vulnerable feature, service, or traffic path is enabled and reachable in the deployed environment. A Cisco access-layer switch can appear in an affected software list without being practically exposed if the vulnerable service is disabled, unreachable, restricted by management ACLs, isolated to a management VLAN, or otherwise mitigated through configuration and architecture. 

What if a Cisco advisory says there is no workaround? 

A no-workaround advisory usually means the vendor is not providing a validated configuration workaround that fully fixes the software defect in the running image. It does not automatically mean every device running that image is exposed, exploitable, or non-compliant. The organization should confirm platform applicability, feature exposure, reachability, mitigation status, and business impact, then remediate the specific affected asset through a fixed release, validated non-affected image, configuration change, isolation, replacement, or documented temporary risk treatment.

Which Cisco vulnerabilities show why exposure matters? 

The IOS XE Web UI vulnerability and Cisco Smart Install are useful examples because the practical exposure depended on enabled and reachable services. For Web UI exposure, disabling HTTP and HTTPS management removed the attack path. For Smart Install exposure, disabling Smart Install with <code>no vstack</code> and restricting management reachability removed or reduced the attack surface. These examples show why access-layer switches should be evaluated by exposure and mitigation status, not IOS version alone. 

What is a configuration attestation memo? 

A configuration attestation memo documents the security controls applied to the access-layer switch environment and maps those controls to the relevant compliance requirements. It can include platform inventory, IOS version, site role, hardened configuration evidence, vulnerability review, compensating controls, risk treatment decisions, logging evidence, and change-management records.</p>