# Free CNX-001 practice questions

> 10 free CNX-001 practice questions with the correct answer and an explanation.

Page: https://devopscodelab.com/exams/comptia-cnx-001/free-practice-questions/
Practice exams: https://devopscodelab.com/exams/comptia-cnx-001/

### Question 1 of 10

A company is deploying 40 branch offices that each need access to workloads in three cloud VPCs/VNets. The network team wants centralized policy, the ability to steer voice traffic over the lowest-latency path, and minimal per-branch configuration. Which architecture BEST fits?

- A. Full-mesh IPsec tunnels from every branch router directly to each of the three cloud virtual gateways
- B. SD-WAN edges at each branch with cloud-hosted SD-WAN gateways in a transit hub that peers with the cloud networks
- C. A dedicated MPLS circuit to each branch with the carrier handing off to a colocation cage that hosts the cloud links
- D. Static GRE tunnels from each branch to the headquarters firewall, which then forwards traffic to the clouds over VPN

**Answer:** B

**Explanation:** SD-WAN provides centralized orchestration, application-aware path selection for voice, and zero-touch provisioning, while a cloud transit hub gives one attachment point for all cloud networks. A full mesh of IPsec tunnels scales poorly (120 tunnels) and has no application steering. MPLS gives predictable paths but is costly per site and offers no dynamic per-application selection. Backhauling through headquarters adds latency and a central bottleneck.

### Question 2 of 10

An architect is planning IP addressing for a cloud environment that will later be connected to an on-premises network that already uses 10.0.0.0/8 extensively. Future mergers are expected. Which approach BEST avoids routing conflicts?

- A. Assign each VPC a public IPv4 range so that internal and external routing never overlap with private addresses
- B. Reserve non-overlapping ranges (an unused RFC 1918 block or 100.64.0.0/10) in a central IPAM first
- C. Use 10.0.0.0/16 in every VPC, because cloud networks are isolated and NAT can be added later if overlaps appear
- D. Use 192.168.0.0/24 for all VPCs because the smaller range limits the number of addresses that can collide

**Answer:** B

**Explanation:** Reserving non-overlapping space in a central IP address management system before deployment prevents conflicts when networks are connected; shared address space or an unused private block is a common choice. Reusing 10.0.0.0/16 everywhere guarantees overlap and forces complex NAT later. Public IPv4 space is scarce and expensive and does not reflect a private design. A /24 everywhere still overlaps across VPCs and is too small for growth.

### Question 3 of 10

A SaaS provider serves users on three continents and wants users routed to the closest healthy regional deployment. If a region fails, traffic must shift automatically within about a minute, and the solution must work at the DNS layer without changing client software. Which design is MOST appropriate?

- A. Anycast of a single public IP to one region with BGP communities that prefer local exits for each continent
- B. Global DNS-based load balancing with latency or geolocation routing, health checks, and low TTL records
- C. A single regional application load balancer with cross-zone load balancing enabled across all availability zones
- D. Round-robin DNS with one A record per region and a TTL of 24 hours to reduce resolver query volume

**Answer:** B

**Explanation:** Global server load balancing at the DNS layer with latency or geolocation policies and health checks answers each user with the nearest healthy region, and low TTLs let changes take effect quickly. A regional load balancer cannot route between continents. Anycasting to a single region gives no regional failover. Round-robin DNS has no health awareness and a 24-hour TTL would keep clients pointed at a failed region.

### Question 4 of 10

A cloud workload in a private subnet must call a third-party payment API. Security requires that outbound traffic be restricted to that API's fully qualified domain name, and the API provider's IP addresses change frequently. Which control BEST enforces this?

- A. Route egress through a cloud firewall that supports FQDN or TLS SNI-based allow rules and deny all other destinations
- B. Attach a NAT gateway to the subnet and rely on the provider allowlisting the NAT gateway's elastic IP address
- C. Use a network ACL on the subnet to deny all outbound traffic except UDP 53 so that only DNS lookups are possible
- D. Add a security group outbound rule allowing TCP 443 to the provider's current IP addresses and update it monthly

**Answer:** A

**Explanation:** An egress firewall with FQDN or SNI-based rules follows the provider's changing addresses and blocks every other destination. Security group rules by IP would break whenever addresses change and need constant maintenance. A NAT gateway provides egress but no destination filtering; the provider's allowlist protects them, not you. Allowing only DNS would block the payment API entirely.

### Question 5 of 10

A network team must prevent unauthorized laptops from gaining access when plugged into conference-room wall jacks, while allowing corporate devices and placing printers on their own VLAN. Which combination BEST achieves this?

- A. Disabling DHCP on the conference-room VLAN and requiring users to request static IP addresses from the help desk
- B. A captive web portal on the conference-room VLAN that asks users to accept an acceptable use policy before browsing
- C. 802.1X EAP-TLS for laptops, MAC Authentication Bypass for printers, and RADIUS-assigned VLANs
- D. Port security limiting each port to one learned MAC address, with sticky learning enabled on every conference-room jack

**Answer:** C

**Explanation:** 802.1X with certificate-based EAP-TLS authenticates corporate devices, MAB handles printers that cannot run a supplicant, and RADIUS can assign each device the right VLAN. Port security with one sticky MAC is easily defeated by spoofing and does not identify devices. Removing DHCP only inconveniences attackers who can set addresses manually. An AUP portal provides no authentication.

### Question 6 of 10

An organization wants network administrators to authenticate to routers and switches with their directory credentials. Auditors require that each command an administrator runs is authorized and logged individually. Which protocol BEST meets this requirement?

- A. Kerberos, because it issues service tickets that the devices can validate for every command entered
- B. LDAPS, because the devices can query the directory directly over TLS and log each bind attempt
- C. TACACS+, because it separates AAA functions and can authorize and log each individual command
- D. RADIUS, because it encrypts the entire packet body and supports per-command authorization by default

**Answer:** C

**Explanation:** TACACS+ separates the AAA functions, encrypts the whole payload, and supports per-command authorization and accounting, which suits device administration. RADIUS combines authentication and authorization, encrypts only the password, and is aimed at network access rather than command authorization. LDAPS can authenticate but provides no command authorization or accounting. Kerberos provides authentication tickets, not command-level authorization on network devices.

### Question 7 of 10 (choose 2)

A network team manages 300 switches with Ansible. Which TWO practices BEST reduce the risk of a bad playbook causing a widespread outage? (Choose two.)

- A. Run the playbook against all 300 switches in parallel so the change window is as short as possible
- B. Roll changes out in small batches with serial execution and automated post-change validation
- C. Store the inventory only on one engineer's laptop so that no one else can run the playbook
- D. Run the playbook in check mode with diff output and review the proposed changes before applying
- E. Disable configuration archiving on the switches to avoid overwriting the last known-good config

**Answer:** B, D

**Explanation:** Check mode with diffs shows exactly what would change before it happens, and serial batches with validation limit blast radius and stop a bad rollout early. Running everywhere at once maximizes impact. Disabling archives removes rollback. Keeping inventory on one laptop creates a single point of failure without reducing change risk.

### Question 8 of 10

A company's DR plan states an RPO of 15 minutes and an RTO of 1 hour for its customer portal, which runs in one cloud region. Which network-related design BEST supports these objectives?

- A. A cold site in a second region where network address ranges are documented and approved for later creation
- B. Synchronous replication across availability zones in the same region and a runbook to rebuild in another region
- C. Async replication to a warm standby region with pre-built networking and health-checked DNS failover
- D. Nightly snapshots copied to a second region, with network resources defined in IaC to be deployed during a disaster

**Answer:** C

**Explanation:** Continuous asynchronous replication keeps data loss within minutes, and a pre-built warm standby with DNS failover can be serving within an hour. Nightly snapshots give up to a 24-hour RPO. Multi-AZ replication does not protect against a regional outage, and a rebuild runbook is too slow. A cold site cannot meet a one-hour RTO.

### Question 9 of 10

A branch's SD-WAN edge reports both underlay links healthy, yet overlay tunnels to the cloud gateway flap every few minutes. Captures show IPsec packets arriving with changing source ports from the branch's broadband ISP. What is the MOST likely cause?

- A. A DNS TTL that is too short for the cloud gateway hostname, causing frequent re-resolution
- B. A duplex mismatch on the LAN-facing switch port connected to the SD-WAN edge appliance
- C. A carrier-grade NAT on the broadband link reassigning ports and expiring the NAT-T mappings
- D. An expired TLS certificate on the SD-WAN orchestrator preventing policy downloads to the edge

**Answer:** C

**Explanation:** Changing source ports on IPsec NAT-T traffic indicate an upstream NAT, often carrier-grade NAT, with short mapping timeouts that break the tunnels; shorter keepalives or a public IP fixes it. An orchestrator certificate would affect control plane, not cause port changes. A LAN duplex mismatch would not change WAN source ports. DNS re-resolution does not alter UDP source ports.

### Question 10 of 10

Users intermittently fail to authenticate to Wi-Fi with 802.1X. RADIUS logs show some requests arrive from the wireless controller but receive no response, and the controller fails over between two RADIUS servers. The second server was recently added. What is the MOST likely cause?

- A. The new RADIUS server lacks a matching client entry or shared secret for the wireless controller
- B. The wireless clients' certificates expired at the same time the second RADIUS server came online
- C. The access points are on a different channel plan than the one the RADIUS server expects to see
- D. The DHCP scope for the wireless VLAN is exhausted, so clients cannot obtain an address after joining

**Answer:** A

**Explanation:** If the new server does not recognize the controller as a client or the shared secret differs, it silently discards requests, causing intermittent failures whenever traffic goes to it. Expired client certificates would fail consistently. RADIUS has no knowledge of channel plans. DHCP exhaustion occurs after successful authentication.

Ready for a timed run? Take the free 15-question CNX-001 mini exam: https://devopscodelab.com/app/exams/comptia-cnx-001/mini/

DevOpsCode Lab is not affiliated with CompTIA. CNX-001 and related names are trademarks of their owners. We write every question ourselves; these are not real exam questions.
