CISSP domain - 13% of the exam

Communication and Network Security

Communication and Network Security is 13% of the CISSP exam. These are the objectives it covers, each with practice questions, with every answer explained.

The domain in numbers

39
Practice questions
3
Objectives
26
Medium
13
Hard

Where people go wrong

  • Swapping IPsec tunnel mode and transport mode

    The wrong answers swap the two IPsec modes or bind a protocol to one of them. Transport mode protects only the original IP payload and leaves the original IP header in place, which suits host-to-host traffic. Tunnel mode encapsulates the entire original IP packet inside a new IP header, hiding the internal addresses on the transit network, which is why gateway-to-gateway links use it. AH and ESP can each operate in either mode, and the difference is structural, not a cipher selection.

    See it in a sample question
  • Offering encryption where isolation or a device check is needed

    The wrong answers offer encryption or authentication where the requirement is isolation or a device check. SD-WAN with IPsec, tunnel-level encryption between hypervisors, and host-based encryption protect data in transit but do not stop lateral movement or separate tenants. Strong authentication and site-to-site VPNs say nothing about the health of the connecting laptop. The correct answers use micro-segmentation, a dedicated VRF, a distributed firewall on each virtual NIC, or posture assessment through network access control.

    See it in a sample question
  • Treating the SDN controller as a trusted component

    The wrong answers treat the SDN controller or the management plane as trusted: an unauthenticated southbound channel, a management plane assumed to be isolated automatically, or a controller on a shared management VLAN with a shared API key that is only rotated or fronted by a firewall. Centralising forwarding decisions makes the controller a high-value target. The correct answers move the control plane to a dedicated out-of-band network with mutually authenticated TLS, and isolate the management plane, because compromising it can rewrite the controls on the data plane.

    See it in a sample question
  • Giving suppliers and contractors broad network reach

    The wrong answers give suppliers and contractors broad reach: staff-network membership, a whole subnet behind a shared-key tunnel, standing VPN access, or the standard VPN with annual self-certification. A compromise at the third party then travels as far as that access does. The correct answers terminate the connection in a dedicated extranet zone or DMZ, allow only the systems and ports needed, use time-bound or certificate credentials, and log sessions. For contractors, a zero trust broker publishes only the required applications.

    See it in a sample question
  • Choosing copper where interference and distance rule it out

    The wrong answers choose copper for a run where interference and distance rule it out. UTP depends on balanced pair geometry, shielded twisted pair keeps the 100 metre channel limit, twinax suits short in-rack links, and coax is not used for modern horizontal runs. Optical fibre carries light through a dielectric core, so electromagnetic fields cannot induce a signal on it, and it supports runs beyond copper limits, including 180 metres beside an MRI suite.

    See it in a sample question

Objectives in this domain

What to study

The official documents this domain's practice questions cite most.

Sample questions from this domain

Free sampleCommunication and Network Securityhard

A network team is comparing IPsec tunnel mode and transport mode for a new design that will protect both site-to-site and host-to-host traffic. Which statement correctly distinguishes the two modes?

  • ATransport mode encapsulates the entire original packet inside a new IP header, while tunnel mode protects only the payload.
  • BTransport mode and tunnel mode are functionally identical and differ only in the cipher suite negotiated during IKE Phase 1.
  • CTransport mode protects only the original IP payload and leaves the original IP header in place, while tunnel mode encapsulates the entire original IP packet inside a new IP header. Correct
  • DTransport mode is used only with AH while tunnel mode is used only with ESP, and they cannot be combined with the other protocol.
Differentiate IPsec transport and tunnel modes by what is protected and whether a new outer IP header is added. Transport mode protects the upper-layer payload using AH or ESP between two endpoints that retain their original IP headers, which suits host-to-host protection. Tunnel mode encapsulates the original IP packet whole inside a new IP header so security gateways can bridge networks, hide internal addressing, and protect headers as well as payload.

Why A is wrong: This swaps the two modes. Candidates who remember that tunnels traverse the public internet sometimes invert the descriptions; tunnel mode is the one that adds a new outer header, and transport mode preserves the original header.

Why B is wrong: IKE negotiates protection algorithms in both modes, but the difference between transport and tunnel is structural (which parts of the original packet are protected and whether a new outer header is added), not a cipher selection, so treating them as identical misses the point.

Why C is correct: Transport mode applies AH or ESP to the payload of the existing packet and is typical for host-to-host protection where endpoints already have routable addresses, while tunnel mode wraps the whole original packet inside a new outer header so gateways can bridge networks with private addressing, which is exactly how the two modes differ.

Why D is wrong: AH and ESP can each operate in either transport or tunnel mode, so binding one protocol to one mode is a tempting simplification but factually wrong and would lead to invalid policy choices in deployment.

Free sampleCommunication and Network Securitymedium

A managed service provider is onboarding a new client whose acceptable use policy permits staff to bring personally owned laptops for work in coffee shops and shared workspaces. The CISO wants to ensure that any endpoint connecting to the corporate virtual desktop environment from those untrusted networks is checked against a current security posture before access is granted, and re-checked whenever posture changes. Which control BEST satisfies that requirement?

  • AA site-to-site IPsec virtual private network anchored at the head office firewall with pre-shared keys.
  • BA network access control solution performing posture assessment with continuous monitoring and dynamic remediation. Correct
  • CA web application firewall in front of the virtual desktop gateway tuned with managed signature rules.
  • DMulti-factor authentication using time-based one-time passwords delivered through an authenticator application.
Use NAC posture assessment with continuous monitoring to gate and re-evaluate endpoint health before and during access to sensitive resources. Network access control combines authentication with endpoint health checks such as patch level, antivirus state, host firewall status, and disk encryption. Pre-admission posture assessment denies or quarantines non-compliant endpoints before they touch protected resources, and continuous monitoring repeats those checks during a session so a device that drops out of compliance is remediated or removed. That combination is the only listed control that addresses both the gated entry and the ongoing posture requirement described.

Why A is wrong: Site-to-site VPNs encrypt traffic between fixed networks and are not designed for mobile user endpoints in coffee shops. They also do not check endpoint posture, so a compromised personal laptop would still reach the virtual desktop environment over the tunnel.

Why B is correct: A NAC platform that performs posture assessment evaluates patch level, antivirus state, disk encryption, and other attributes before granting access, and continuous monitoring re-checks posture during the session. Dynamic remediation can quarantine an endpoint that drifts out of compliance, which is exactly the gated and ongoing enforcement the CISO has asked for, making it the BEST fit.

Why C is wrong: A WAF inspects HTTP traffic for application-layer attacks but has no view of the connecting endpoint's patch level, antivirus status, or encryption state. It addresses a different threat class and cannot enforce the posture-gated access the requirement specifies.

Why D is wrong: Strong authentication confirms who the user is but says nothing about whether the device they are using is healthy. A malware-laden personal laptop with valid credentials and a correct TOTP would still satisfy MFA, so it does not meet the posture-checking requirement on its own.

Free sampleCommunication and Network Securityhard

An architect is documenting a software-defined networking deployment for the audit committee. Which statement most accurately describes the security-relevant property of the SDN control plane and data plane separation?

  • ASeparating control and data planes removes the need for authentication on the southbound interface because controller commands are trusted by default.
  • BSeparating control and data planes centralises policy decisions and makes the controller a high-value target whose compromise can reprogram forwarding behaviour fabric-wide. Correct
  • CSeparating control and data planes guarantees that forwarding decisions are made locally at each switch, eliminating any centralised choke point.
  • DSeparating control and data planes mainly improves cable management and has no meaningful effect on the network's trust boundaries.
Recognise that SDN concentrates policy in a controller, making controller and southbound channel security central to fabric trust. In SDN, switches forward according to flow rules pushed from a controller via the southbound interface, while northbound APIs expose policy to applications. Because the controller can rewrite forwarding behaviour everywhere, its authentication, integrity, and channel protection are the primary security concerns of the design, not a side effect of refactoring control logic.

Why A is wrong: Some candidates assume that an internal fabric link is inherently trusted, but the southbound channel (for example OpenFlow) must be authenticated and ideally TLS-protected; treating it as trusted is exactly the misconfiguration that lets an attacker on the management network rewrite flow tables.

Why B is correct: SDN moves the decision logic into a logically centralised controller and leaves switches as policy enforcers, which simplifies orchestration but concentrates risk: an attacker who subverts the controller or the southbound channel can install forwarding rules across the fabric, so the controller becomes the most sensitive trust anchor in the design.

Why C is wrong: This reverses the architecture: SDN deliberately moves decision making off the switch into a controller, so claiming local decisions and no central choke point describes traditional distributed networking, not SDN.

Why D is wrong: Framing the change as a cabling or operational matter understates the security implication: the controller becomes a new privileged component with its own attack surface, so dismissing the trust-boundary shift is the wrong answer in a security review.

Free sampleCommunication and Network Securitymedium

A healthcare provider is onboarding a transcription supplier that will pull dictation files from an internal SFTP server. The security manager wants to limit the supplier's reach if the supplier's network is breached. Which control set BEST reflects a risk-led approach to this third-party connection?

  • AAllow the supplier outbound to the entire internal subnet through an IPsec tunnel using a shared pre-shared key.
  • BTerminate a site-to-site IPsec tunnel in a dedicated DMZ, restrict firewall rules to the SFTP host and port, and require certificate-based authentication. Correct
  • CPublish the SFTP service on the public internet and rely on strong supplier-side passwords and IP allow-listing for protection.
  • DProvision a jump host inside the production network that the supplier reaches over an SSL VPN with username and password authentication.
Constrain third-party connectivity using segmentation, least-privilege firewalling, and strong authentication. Third-party connections fail when blast radius is not pre-contained. Terminating the tunnel in a DMZ keeps the supplier away from production by default, host and port restrictions express least privilege at the firewall, and certificate-based peer authentication removes the shared-secret weakness that is common in supplier compromise scenarios.

Why A is wrong: A tunnel with a shared key would establish encrypted transport but granting access to the whole subnet violates least privilege and amplifies blast radius if the supplier is compromised.

Why B is correct: Terminating in a DMZ limits exposure if the tunnel is abused, port and host restrictions enforce least privilege, and certificate-based authentication removes the shared-secret risk common in supplier breaches.

Why C is wrong: Public exposure with passwords and IP allow-listing is operationally simple but leaves the service reachable by anyone scanning the internet and assumes the supplier's egress IP is stable and uncompromised.

Why D is wrong: A jump host adds a useful pivot point but placing it in production rather than a DMZ, combined with single-factor authentication, undermines the containment objective.

Free sampleCommunication and Network Securitymedium

A cabling contractor is replacing horizontal runs between an industrial floor and a wiring closet. The path crosses several large variable-frequency drives that produce strong electromagnetic interference, and the runs occasionally exceed 90 metres. Which transmission medium choice BEST addresses both the EMI exposure and the distance characteristic of the route?

  • AUnshielded twisted pair Category 6, because its tighter twist rate cancels the surrounding interference and supports gigabit speeds in the typical office channel.
  • BMultimode optical fibre, because the dielectric core is immune to electromagnetic interference and supports runs well beyond copper distance limits. Correct
  • CShielded twisted pair Category 6A with a drain wire bonded to ground at the patch panel, because the foil contains coupled interference within the cable jacket.
  • DCoaxial cable with a braided shield, because the concentric geometry provides constant impedance and the shield rejects radiated interference.
Recognise that optical fibre is the appropriate transmission medium when both strong EMI and copper distance limits are concerns. Copper transmission media carry signals as electrical currents and are therefore susceptible to electromagnetic induction from nearby motors, drives, and unshielded power runs. Shielding reduces but does not eliminate this coupling, and the 100 metre balanced cabling limit still applies. Optical fibre transmits modulated light through a glass core, which is non-conductive and unaffected by external electromagnetic fields, and multimode fibre supports distances well beyond copper limits, making it the correct choice when both EMI immunity and reach are needed.

Why A is wrong: UTP relies on balanced pair geometry rather than shielding, so strong external EMI still couples into the run and the 100 metre channel limit is unchanged. It is a tempting choice because Category 6 is common, but it does not address the dominant risk.

Why B is correct: Optical fibre carries light through a glass dielectric, so external electromagnetic fields cannot induce a signal on the conductor in the way they do on copper. Multimode fibre also comfortably exceeds the 100 metre copper channel boundary, addressing both constraints in the scenario.

Why C is wrong: STP improves EMI immunity over UTP and is plausible for noisy environments, but it still carries the same 100 metre copper channel limit and requires careful bonding to avoid ground loops that can themselves inject noise.

Why D is wrong: Coax does reject some EMI through its shield and was historically used for noisy environments, but modern structured cabling does not use coax for Ethernet horizontal runs, and the medium does not give the distance advantage the route requires.

Other domains in this exam

See also the CISSP cert hub, the study guide, and the cheat sheet.

Examworthy is not affiliated with or endorsed by ISC2. Original, blueprint-aligned practice material only.