Past ExamsRwanda ICT preparation

Network Administrator practice questions with answers

Updated 15 September 2026 · 2 min read

The questions below are chosen for network administrator work: addressing, switching, routing, services and security, in roughly the balance the job uses them. Answer each one before you open the explanation. Guessing and then reading the answer feels productive, but it teaches far less than committing to an answer and finding out why it was wrong.

How this guide was made

Written for this site with AI assistance, then checked automatically before publishing. The topic table, sample questions, worked examples and study plan are generated from this site’s question bank: each worked example is recalculated by separate code, and each sample answer says where it came from.

The guide deliberately states no salaries, grades, pass marks or exam dates. Those change, and they could not be checked here, so take them from the advert itself.

This site is independent and is not affiliated with MIFOTRA or any employer.

Practice questions

  1. Users report that their web browsers are failing to negotiate TLS encryption with a newly deployed internal server. Which OSI layer is failing to establish this secure connection?

    1. Presentation
    2. Session
    3. Application
    4. Transport
    Show the answer

    A. Presentation

    Objective 1.1: The Presentation layer (Layer 6) handles encryption/decryption like TLS. Application handles HTTP, Session handles teardown, Transport handles TCP.

    Original practice question, written against published exam objectives. Open this question

  2. Which cable type is most commonly used for Ethernet connections?

    1. Fiber optic cable
    2. Twisted pair cable
    3. Serial cable
    4. Coaxial cable
    Show the answer

    B. Twisted pair cable

    Twisted pair (Cat5e/Cat6) is by far the most common Ethernet medium. Fibre is used for long or high-capacity links, not for the typical desk connection.

    Real past-paper question; answer cross-checked against the original session. Open this question

  3. An enterprise connects its 5 branch offices directly to every other branch office using point-to-point links for maximum redundancy. What topology does this represent?

    1. Ring
    2. Partial Mesh
    3. Full Mesh
    4. Star
    Show the answer

    C. Full Mesh

    Objective 1.2: A full mesh topology connects every node to every other node. Partial mesh only connects some, Star uses a central hub.

    Original practice question, written against published exam objectives. Open this question

  4. During a network outage, a technician finds that a fiber optic cable was accidentally severed. Which layer of the OSI model is directly impacted by this physical break?

    1. Network
    2. Application
    3. Data Link
    4. Physical
    Show the answer

    D. Physical

    Objective 1.1: The Physical layer (Layer 1) encompasses the physical media like fiber cables. Network is logical routing, Data Link is MAC.

    Original practice question, written against published exam objectives. Open this question

  5. A packet is destined for 192.168.1.130. The routing table has 192.168.1.0/24 (AD 110) and 192.168.1.128/26 (AD 120). Which path does the router choose?

    1. It load balances
    2. The 192.168.1.128/26 route
    3. The 192.168.1.0/24 route
    4. It drops the packet
    Show the answer

    B. The 192.168.1.128/26 route

    The router uses the longest prefix match rule first, regardless of administrative distance. The /26 is more specific.

    Original practice question, written against published exam objectives. Open this question

  6. A network architect is designing a multi-area OSPF network. What is the strict requirement regarding Area 0?

    1. Area 0 must contain the ASBR
    2. All other areas must connect directly to Area 0
    3. Area 0 cannot contain summary LSAs
    4. Area 0 must use authentication
    Show the answer

    B. All other areas must connect directly to Area 0

    In OSPF, Area 0 is the backbone. All non-backbone areas must physically (or logically via virtual link) connect directly to Area 0 to prevent routing loops.

    Original practice question, written against published exam objectives. Open this question

  7. You are troubleshooting an IPsec VPN tunnel where ESP packets are being dropped by an intermediary firewall. Which OSI layer protocols are being blocked?

    1. Session
    2. Data Link
    3. Network
    4. Transport
    Show the answer

    C. Network

    Objective 1.1: IPsec (including ESP) operates at the Network layer (Layer 3). Transport is Layer 4, Session Layer 5, Data link Layer 2.

    Original practice question, written against published exam objectives. Open this question

  8. An endpoint detection system flags unusual PowerShell activity. Investigation reveals a malicious script running entirely in RAM, heavily obfuscated, leaving no executable footprint on the hard drive. This is...

    1. Rootkit
    2. Logic bomb
    3. Fileless malware
    4. Trojan
    Show the answer

    C. Fileless malware

    Fileless malware operates in memory (RAM) and abuses native legitimate tools (like PowerShell or WMI) rather than dropping an executable file on the disk.

    Original practice question, written against published exam objectives. Open this question

How to get the most from them

Cover the answer and write down your choice with a one-line reason. Then open the explanation and compare your reason with it, not just your letter. A right answer for a wrong reason is a mistake that simply has not cost you marks yet.

When you get one wrong, decide which kind of mistake it was. Did you not know the fact, misread the stem, slip in a calculation, or talk yourself out of the right option? Each kind needs a different fix, and most candidates find that one kind dominates.

What comes after this set

The two calculations further down are generated fresh for this page and checked by separate code, so use them to test your subnetting speed. The questions from the full bank are real items on the same areas. You can read all of them; their answers open with an access code.

Two calculations to try

Try each one on paper first. Every figure is generated and then recalculated by separate code before it is published.

  1. Subnet details for a /28 address

    A host is configured as 192.168.163.169/28. Find the subnet mask, network address, broadcast address, usable host range and number of usable hosts.

    Show the working
    1. /28 means 28 network bits, so the mask is 255.255.255.240.
    2. The interesting octet is the fourth, where the mask is 240. Block size = 256 − 240 = 16.
    3. The network address is the multiple of 16 at or below the address in that octet: 192.168.163.160.
    4. The broadcast is one below the next network: 192.168.163.175.
    5. Usable hosts run from 192.168.163.161 to 192.168.163.174: 2^4 − 2 = 14 addresses.

    Answer: Mask 255.255.255.240, network 192.168.163.160, broadcast 192.168.163.175, hosts 192.168.163.161–192.168.163.174 (14 usable)

  2. Choosing the smallest subnet that fits

    A department needs 12 usable IPv4 host addresses on one subnet. What is the longest prefix (smallest subnet) that fits, and how many usable addresses does it give?

    Show the working
    1. Usable hosts = 2^(host bits) − 2, because the network and broadcast addresses cannot be assigned.
    2. Find the fewest host bits h where 2^h − 2 ≥ 12: h = 4 gives 14, while h = 3 gives only 6.
    3. Prefix = 32 − 4 = /28.

    Answer: /28, giving 14 usable addresses

More from the full bank

Anyone can read these questions. Their answers and explanations open with an access code.

Common questions

Are these the real exam questions?

Nobody outside the examining body has the questions for a future test. These are practice items written against the same knowledge, alongside real past-paper questions where this site holds them.

How many practice questions should I do before the exam?

Enough that your error log stops growing in the same areas. For most people that means several hundred questions spread over a few weeks, with mistakes revisited, rather than one enormous batch.

Should I time myself on practice questions?

Not at first. Work untimed until you are accurate, then add a clock. Speed built on shaky accuracy tends to collapse under exam pressure.

More Network Administrator guides

The same guide for related positions