ISC2-CC domain - 24% of the exam

Network Security

Network Security is 24% of the ISC2-CC exam. These are the objectives it covers, each with practice questions, with every answer explained.

The domain in numbers

77
Practice questions
3
Objectives
47
Medium
30
Hard

Where people go wrong

  • Treating an IDS as able to block traffic

    The wrong answers give an intrusion detection system the power to stop an attack, or treat an inline prevention system as a passive alarm. An IDS sits out of band on a mirror port, sees copies of packets and raises alerts, so a malicious request such as a SQL injection attempt still completes. An IPS sits inline in the traffic path and can drop or reset matching packets in real time. A SIEM and a honeypot are further wrong options: neither inspects live traffic to block it.

    See it in a sample question
  • Expecting a stateless filter to allow return traffic

    The wrong answers reach for a stateless packet filter, a mirrored access list, a screened subnet, a proxy or a block of open high ports when replies to internal connections must be let back in. A stateless filter keeps no memory of sessions, so it cannot tell a reply from an unsolicited inbound packet, and its only workaround is a broad static rule such as every TCP port above 1023. A stateful firewall records each outbound connection in a state table and admits only matching return traffic.

    See it in a sample question
  • Confusing site-to-site and remote-access VPNs

    The wrong answers offer a site-to-site VPN where one roaming laptop needs access, or a control that never creates a tunnel at all, such as TLS on the public website, WPA2 or port forwarding. A site-to-site VPN joins two fixed networks through gateways, so individual users on changing addresses cannot be onboarded through it. A remote-access VPN runs client software on the device and builds an encrypted tunnel to the office gateway, protecting traffic across untrusted cafe and hotel networks.

    See it in a sample question
  • Keeping implicit trust for traffic inside the network

    The wrong answers keep trusting a request because of where it comes from: stacked perimeter firewalls, a single strong boundary in a castle-and-moat design, or NAT. Each leaves the inside implicitly trusted, so an attacker who reaches the corporate LAN moves freely. Zero trust removes that assumption. No user or device is trusted by default, and every request is authenticated and authorised regardless of network location, with least-privilege access granted per session.

    See it in a sample question
  • Placing Layer 2 and Layer 3 functions at the wrong layer

    The wrong answers pick the neighbouring layer. Layer 2 is chosen for logical addressing and routing, and Layer 3 for MAC-based forwarding, because both involve addressing. The data link layer uses MAC addresses within one local segment, which is where a switch makes its forwarding decision. The network layer uses logical IP addresses to route between separate networks, so an address-based packet filter operates at Layer 3. Layer 4 handles ports and segments instead.

    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 sampleNetwork Securityhard

A team debates the essential difference between an intrusion detection system and an intrusion prevention system protecting the same segment. Which statement most accurately captures the distinction?

  • AAn IDS is deployed inline and drops malicious packets, whereas an IPS only copies traffic to review it after the fact.
  • BAn IDS decrypts TLS sessions by default, whereas an IPS is unable to inspect any encrypted traffic at all.
  • CAn IDS can operate only at the host level, whereas an IPS can operate only at the network level.
  • DAn IDS monitors and alerts on suspicious traffic out of band, whereas an IPS sits inline and can block the traffic in real time. Correct
An IDS passively detects and alerts on suspicious traffic, while an IPS sits inline and can actively block it. An IDS typically receives a copy of traffic and raises alerts without affecting the flow, whereas an IPS is positioned in the path so it can inspect and then drop or reset offending traffic before it reaches the target.

Why A is wrong: Tempting because it names inline placement and blocking, but it reverses the two: dropping inline traffic is the IPS role, not the IDS role.

Why B is wrong: Tempting because encryption does complicate inspection, but neither device decrypts TLS by default, and the ability to inspect encrypted traffic depends on configuration, not on the IDS or IPS distinction.

Why C is wrong: Tempting because host and network variants exist, but both IDS and IPS come in host-based and network-based forms, so neither is restricted to a single level.

Why D is correct: Correct: the defining difference is that detection systems observe and alert passively, while prevention systems are placed in the traffic path and can actively drop or block.

Free sampleNetwork Securityhard

A network engineer explains that one type of firewall records each permitted outbound connection in a table so that the matching return traffic is allowed back automatically without a separate inbound rule. Which firewall characteristic does this describe?

  • AStateless packet filtering, which evaluates every packet in isolation against the static rule set with no memory of prior traffic.
  • BApplication-layer proxying, which terminates each session and rebuilds a fresh connection to the destination on the client's behalf.
  • CStateful inspection, which tracks the state of active connections and permits return traffic that matches an existing table entry. Correct
  • DStatic network address translation, which maps one fixed internal address to one fixed external address at the boundary.
A stateful firewall tracks active connections in a state table so that matching return traffic is permitted without an explicit inbound rule. Stateful inspection stores the source, destination, ports and connection state of each permitted flow, then matches inbound replies against that table, which is why the corresponding return packets are allowed automatically.

Why A is wrong: Tempting because packet filters also use rules, but a stateless filter keeps no connection table, so return traffic needs its own explicit rule rather than being allowed automatically.

Why B is wrong: Tempting because proxies are also connection-aware, but proxying works by mediating full sessions at Layer 7, not by adding return traffic to a state table at the transport layer.

Why C is correct: Correct: a stateful firewall records established sessions in a state table and uses that entry to allow the corresponding reply traffic without a separate rule.

Why D is wrong: Tempting because NAT also keeps mappings, but static NAT rewrites addresses and does not decide whether return traffic is permitted based on connection state.

Free sampleNetwork Securityhard

Elm Consulting has staff working from home who each need encrypted access to internal file servers across the public internet from their personal laptops. Which solution is designed for this access pattern?

  • AA site-to-site IPsec tunnel established between two fixed office routers
  • BEnabling TLS on the company public marketing website only
  • CForwarding the file server port through the edge router to the public internet
  • DA remote-access VPN client on each laptop terminating at the office VPN gateway Correct
A remote-access VPN gives individual roaming users an encrypted, authenticated tunnel to internal resources over the internet. A remote-access VPN terminates a per-user encrypted tunnel at the office gateway, authenticating each home worker and protecting traffic in transit, which a site-to-site tunnel between fixed sites or plain website TLS cannot do for individual laptops.

Why A is wrong: A site-to-site tunnel is tempting because it also uses IPsec, but it joins two fixed networks rather than serving individual roaming laptops, so home workers on changing addresses cannot be onboarded through it.

Why B is wrong: TLS on the public website secures visits to that site, but it provides no encrypted channel to the internal file servers the staff must reach, so it does not solve the remote access requirement.

Why C is wrong: Port forwarding would make the file server reachable, but it exposes it directly to the internet without an encrypted, authenticated tunnel, creating a serious risk rather than providing secure remote access.

Why D is correct: A remote-access VPN authenticates each roaming user and builds an encrypted tunnel from their laptop to the office gateway, giving individual home workers protected reach to internal file servers over the public internet.

Free sampleNetwork Securityhard

After a breach, Ballarat Freight adopts a model in which no device is trusted simply for being on the corporate LAN. Every request to an application must be authenticated and authorised, and access is granted per session with least privilege. Which principle BEST describes this approach?

  • ADefence in depth, achieved by stacking several perimeter firewalls in series.
  • BPerimeter security, where the internal network is trusted once traffic passes a hardened boundary.
  • CZero trust, which treats every request as untrusted and verifies it regardless of the requester's network location. Correct
  • DNetwork address translation, which hides internal host addresses from external observers.
Zero trust verifies and authorises every request per session with least privilege, rather than trusting a device for being inside the network perimeter. Zero trust replaces implicit network-location trust with continuous verification: each request is authenticated and authorised and access is scoped to the minimum needed for that session, which matches the described post-breach model.

Why A is wrong: Layering defences is sound, but chaining perimeter firewalls still assumes the inside is trusted, which is the assumption the described model deliberately removes.

Why B is wrong: This castle-and-moat approach grants implicit internal trust, the opposite of verifying every request regardless of location.

Why C is correct: Zero trust removes implicit trust based on network position and verifies identity and authorisation for each request, granting least-privilege access per session.

Why D is wrong: NAT conserves and masks addresses but makes no authentication or authorisation decision, so it does not describe a trust model at all.

Free sampleNetwork Securitymedium

A trainee is learning the OSI reference model and asks which layer is responsible for logical addressing and for selecting a path between networks so that a packet can reach a host on a different subnet. Which layer performs this function?

  • AThe data link layer (Layer 2), which frames bits and uses MAC addresses to move data between devices on the same physical segment.
  • BThe network layer (Layer 3), which uses logical IP addresses and routing decisions to forward packets between separate networks. Correct
  • CThe transport layer (Layer 4), which segments data and provides reliable end-to-end delivery using port numbers to identify applications.
  • DThe session layer (Layer 5), which establishes, manages and tears down dialogues between two communicating applications.
The OSI network layer (Layer 3) provides logical IP addressing and routing to move packets between separate networks. Routing between subnets requires logical addresses that are independent of hardware, which is exactly what the network layer supplies through IP addressing and forwarding decisions, while lower layers only move data within one segment.

Why A is wrong: This is tempting because Layer 2 also does addressing, but it uses hardware MAC addresses and only moves frames within a single local segment, not between separate networks.

Why B is correct: Correct: Layer 3 handles logical addressing with IP and makes routing decisions, allowing a packet to travel from one subnet to another.

Why C is wrong: This is tempting because Layer 4 does end-to-end delivery, but it identifies applications with ports and does not perform logical network addressing or path selection.

Why D is wrong: This is tempting as it sits near the middle of the stack, but Layer 5 manages conversations between applications and plays no role in addressing or routing.

Other domains in this exam

See also the ISC2-CC cert hub, the study guide, and the cheat sheet.

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