Securing Enterprise Infrastructure
Security+ SY0-701 · Security Architecture
Designing a secure network
Objective 3.2 is scenario-based: the exam describes a network requirement and asks which design or control best meets it. The underlying principles are consistent — defense in depth, least privilege, and segmentation — and the scenarios test whether you can apply them to concrete infrastructure choices. Think of the network as a series of layers, each with its own controls, so that no single failure exposes everything.
Network segmentation and zones
Segmentation divides a network into separate zones with controlled traffic between them. A VLAN (virtual local area network) logically separates traffic on the same physical switches, so the finance VLAN cannot see the guest VLAN. A DMZ (demilitarized zone) is a buffer network between the internet and the internal network, hosting public-facing servers such as web servers and mail servers; if a DMZ server is compromised, the attacker is still outside the internal network. An air gap is complete physical separation with no network connection at all, used for the most sensitive systems. Jump servers (bastion hosts) are hardened systems that provide the only administrative access path into a protected zone: administrators log in to the jump server first, and from there reach internal systems. Microsegmentation takes the idea further by isolating individual workloads, often in software-defined networks, so that even systems in the same VLAN cannot talk to each other unless a policy allows it.
Firewalls, proxies, and load balancers
A firewall filters traffic based on rules. A stateful firewall tracks the state of connections and allows return traffic for sessions that were initiated from inside. A next-generation firewall (NGFW) adds application awareness, intrusion prevention, and user identity to its decisions. A web application firewall (WAF) sits in front of web servers and inspects HTTP traffic for attacks such as SQL injection and XSS. Firewall rules should follow an implicit deny: block everything, then allow only what is needed.
A proxy server sits between users and the internet, forwarding requests on their behalf; a forward proxy handles outbound traffic and can filter content, while a reverse proxy sits in front of servers and hides them from direct internet access. A load balancer distributes traffic across multiple servers, improving both performance and availability — if one server fails, the others take over. Port security on switches limits which devices can connect to a port, and 802.1X is the standard for port-based network access control, requiring a device to authenticate before it can join the network.
Secure protocols and remote access
The exam expects you to know which protocol to use for each job, and which legacy protocols to avoid. For remote administration, use SSH instead of Telnet (which sends credentials in clear text). For file transfer, use SFTP or FTPS instead of FTP. For web traffic, use HTTPS (TLS) instead of HTTP. For email, use SMTPS/IMAPS/POP3S (TLS-encrypted) instead of plain SMTP/IMAP/POP3. For directory services, use LDAPS instead of LDAP. For DNS, DNSSEC adds authentication to DNS responses, preventing poisoning. For network management, use SNMPv3 (which supports authentication and encryption) instead of SNMPv1 or v2c, which use community strings sent in clear text.
A VPN (virtual private network) creates an encrypted tunnel over an untrusted network. IPSec operates at the network layer and can run in tunnel mode (encrypting the whole packet, used for site-to-site VPNs) or transport mode (encrypting only the payload, used for host-to-host communication). SSL/TLS VPNs operate at the application layer and are commonly used for remote user access through a web browser. A split tunnel lets VPN users send some traffic through the tunnel and some directly to the internet — convenient, but a security risk because internet-bound traffic bypasses corporate controls; a full tunnel sends all traffic through the VPN.
Site and infrastructure considerations
Physical infrastructure matters too. Redundancy eliminates single points of failure: redundant power supplies, network links, and internet connections. Load balancing and clustering keep services available when a component fails. Hot sites, warm sites, and cold sites are alternate locations for disaster recovery, in descending order of readiness and cost (covered in depth in the resilience guide). HVAC, fire suppression, and uninterruptible power supplies (UPS) protect the physical environment. And cable management and secure wiring closets prevent unauthorized physical access to the network itself.
When the exam gives you a scenario, match the requirement to the control: public-facing servers go in a DMZ; administrative access goes through a jump server; untrusted networks get a VPN; legacy clear-text protocols get replaced with encrypted versions; and a single point of failure gets redundancy. The correct answer is usually the one that adds a layer of defense without breaking the business requirement.
Common ports and protocols
The exam expects you to recognize the well-known ports for common services, because firewall and log scenarios refer to them constantly. Memorize this table:
| Port | Protocol | Use |
|---|---|---|
| 21 | FTP | File transfer (insecure — use SFTP/FTPS) |
| 22 | SSH / SFTP / SCP | Secure remote administration and file transfer |
| 23 | Telnet | Remote administration (insecure — use SSH) |
| 25 | SMTP | Sending email |
| 53 | DNS | Domain name resolution |
| 80 | HTTP | Web traffic (insecure — use HTTPS) |
| 110 | POP3 | Retrieving email (use POP3S, port 995) |
| 143 | IMAP | Accessing email (use IMAPS, port 993) |
| 389 | LDAP | Directory queries (use LDAPS, port 636) |
| 443 | HTTPS | Secure web traffic (TLS) |
| 445 | SMB | Windows file sharing |
| 3389 | RDP | Remote Desktop Protocol |
When a scenario mentions traffic on an unexpected port — for example, outbound traffic on port 53 to an unknown server, or a workstation accepting connections on port 3389 — that is often the clue that something is wrong. Port 53 traffic to an unfamiliar DNS server can indicate DNS tunneling (smuggling data out inside DNS queries), and an open RDP port on a workstation can indicate a backdoor. The port number is the exam's way of telling you which service is involved without naming it.
One final infrastructure concept is high availability through clustering: two or more servers configured so that if one fails, another takes over automatically. Active-active clustering runs all nodes serving traffic at once, while active-passive keeps a standby node idle until the primary fails. Combined with load balancing and redundant network paths, clustering is how organizations meet demanding availability requirements — and the exam may ask which design keeps a service running when a single server fails.
Exam traps to watch for
- Putting public servers on the internal network. Web and mail servers that face the internet belong in a DMZ, not inside the trusted network.
- Choosing a legacy clear-text protocol. Telnet, FTP, HTTP, and SNMPv1/v2c send data (sometimes including credentials) unencrypted — always pick the secure version.
- Confusing tunnel mode with transport mode. IPSec tunnel mode encrypts the whole packet (site-to-site); transport mode encrypts only the payload (host-to-host).
- Thinking a split tunnel is more secure. A split tunnel lets some traffic bypass the VPN and corporate controls; a full tunnel sends everything through the VPN.
Practice questions
Test yourself on this guide. Answers are checked instantly, with a full explanation for each question.
Q1.A company wants to host a public web server that customers can reach from the internet, but it must not allow internet traffic to reach the internal network if the web server is compromised. Where should the web server be placed?
Explanation: A DMZ (demilitarized zone) is a buffer network between the internet and the internal network; public-facing servers live there so that a compromise of the web server does not give the attacker access to internal systems. Option A is wrong because a strong password does not isolate the server from the internal network. Option B is wrong because placing a public server on the database VLAN exposes the most sensitive systems. Option D is wrong because a jump server provides administrative access into a protected zone; it is not a placement for public servers.
Exam objective: SY0-701 Objective 3.2: Given a scenario, apply security principles to secure enterprise infrastructure
Q2.A system administrator needs to manage a Linux server remotely. The current method uses Telnet. Which change best secures this remote administration?
Explanation: SSH encrypts the entire remote administration session, including the login credentials, whereas Telnet sends everything in clear text. Option D is wrong because a longer password is still sent in clear text over Telnet. Option B is wrong because FTP is a file transfer protocol, not a remote administration protocol, and it also sends credentials in clear text. Option A is wrong because it is an impractical overreaction when a secure protocol solves the problem.
Exam objective: SY0-701 Objective 3.2: Given a scenario, apply security principles to secure enterprise infrastructure
Q3.Two branch offices need a permanent, encrypted connection between their networks so that traffic between offices is protected over the internet. Which solution should be used?
Explanation: A site-to-site IPSec VPN in tunnel mode encrypts entire packets between the two office networks, providing a permanent encrypted connection for all inter-office traffic. Option C is wrong because an SSL VPN with split tunneling is for individual remote users, not for connecting two networks, and split tunneling would let some traffic bypass the tunnel. Option A is wrong because a reverse proxy hides servers from direct internet access but does not encrypt traffic between offices. Option D is wrong because a WAF inspects web traffic for attacks; it does not create an encrypted tunnel.
Exam objective: SY0-701 Objective 3.2: Given a scenario, apply security principles to secure enterprise infrastructure
Q4.A company wants to prevent employees from accessing known malicious websites. Which control blocks access to known malicious domains?
Explanation: DNS filtering blocks access to known malicious domains by preventing their names from resolving, stopping users from reaching dangerous sites. Option C is wrong because a reverse proxy sits in front of servers to handle inbound traffic, not outbound user requests. Option D is wrong because a load balancer distributes traffic across servers. Option B is wrong because a jump server provides administrative access to protected systems.
Exam objective: SY0-701 Objective 3.2: Given a scenario, apply security principles to secure enterprise infrastructure
Q5.An organization is concerned that an attacker could plug an unauthorized laptop into an unused Ethernet port in a meeting room and gain network access. Which control prevents this?
Explanation: Port security combined with 802.1X requires a device to authenticate before the switch port allows it onto the network, so an unauthorized laptop plugged into a wall port gets no access. Option D is wrong because a WAF inspects web application traffic, not physical port access. Option C is wrong because DNSSEC authenticates DNS responses. Option A is wrong because full-disk encryption protects data on a stolen drive, not network access.
Exam objective: SY0-701 Objective 3.2: Given a scenario, apply security principles to secure enterprise infrastructure
Q6.A web application is under attack from SQL injection attempts. The application code cannot be changed quickly. Which control can inspect incoming HTTP requests and block injection attempts before they reach the application?
Explanation: A web application firewall (WAF) inspects HTTP traffic at the application layer and can block SQL injection, XSS, and other web attacks before they reach the application. Option D is wrong because a stateful firewall filters traffic by IP address, port, and connection state; it does not understand HTTP content. Option C is wrong because a load balancer distributes traffic across servers. Option A is wrong because a forward proxy forwards outbound user requests rather than protecting inbound application traffic.
Exam objective: SY0-701 Objective 3.2: Given a scenario, apply security principles to secure enterprise infrastructure
Q7.A network engineer is writing firewall rules for a new server. Which rule order follows security best practice?
Explanation: Best practice is implicit deny: block everything by default and then allow only the specific traffic the server requires (least privilege applied to network access). Option A is wrong because allowing everything and blocking known-bad addresses leaves the server open to unknown threats. Option B is wrong because allowing all common ports exposes unnecessary services. Option C is wrong because time-based rules do not reduce the attack surface during business hours, when attacks still happen.
Exam objective: SY0-701 Objective 3.2: Given a scenario, apply security principles to secure enterprise infrastructure
Q8.A company wants its public website to stay available even if one of its web servers fails. Which solution distributes incoming traffic across multiple servers so the failure of one does not take the site down?
Explanation: A load balancer distributes incoming traffic across multiple servers, so if one server fails the others continue serving the website. Option C is wrong because a jump server provides administrative access to protected systems. Option B is wrong because a honeypot is a decoy system used to detect attackers. Option D is wrong because DNSSEC authenticates DNS responses and does not distribute traffic.
Exam objective: SY0-701 Objective 3.2: Given a scenario, apply security principles to secure enterprise infrastructure