More study resources

About this exam

The Certified Unix Network Professional (CUNP) is a vendor-neutral credential aimed at administrators who already run Unix-like systems day to day and now need to prove they can design, configure, secure, and debug the network layer on those systems. It suits mid-level sysadmins, NOC and infrastructure engineers, and DevOps practitioners whose work spans Linux or BSD hosts, daemons, packet filters, and routing tables. The exam is multiple choice and tests applied judgement rather than trivia: reading interface state and route output, reasoning about the TCP/IP stack and where a failure sits in it, configuring services and daemons, writing firewall and traffic-shaping rules, applying authentication and access controls, diagnosing with monitoring and packet tools, and automating checks in shell. Expect scenario questions where several answers look plausible and the deciding detail is a flag, a file location, or the order of operations. Preparation works best hands-on: build a small lab with two or three hosts or namespaces, break things deliberately, and practise explaining each symptom to yourself. Pair that with the man pages and vendor documentation for the tools you actually use. Confirm current exam logistics, including the number of questions, time allowed, fee, passing score, retake rules, and validity period, directly with the certifying body, since it publishes the authoritative figures and this page does not restate them.

What you'll study

Common questions

About CUNP preparation

What is the exam format?

It is delivered as multiple-choice questions covering the eight published domains, weighted from fundamentals and protocols through configuration, services, routing and firewalling, security, troubleshooting, and scripting. There is no hands-on lab component in the stated format, so you must be able to predict command output and diagnose scenarios on paper. Check the certifying body's site for the current delivery method.

How hard is CUNP?

Candidates with a year or two of real Unix administration usually find the fundamentals and service domains comfortable and struggle with routing, traffic control, and firewall rule ordering. The difficulty is applied reasoning, not memorisation: questions often hinge on one flag or the sequence in which rules and routes are evaluated. Lab practice narrows the gap quickly.

How many questions and how long do I get?

The certifying body publishes the current question count and time limit, and those figures change between exam versions, so confirm them on its official site before you book. As a planning habit, budget roughly a minute per question plus review time, and practise pacing with timed sets of mixed-domain questions.

What does it cost and what is the passing score?

Fees and the passing scaled score are set by the certifying organisation and are not restated here. Check its official website for the current price, any regional variation, and the score report format. Do not rely on forum posts, which often quote outdated numbers from earlier exam versions.

What happens if I fail, and how long is the certification valid?

Retake policy and recertification or validity periods are defined by the certifying body and can differ by region and exam version. Look up the current retake waiting period, any fee for subsequent attempts, and the validity term on the official site. Plan your first attempt when your practice scores are consistently comfortable rather than gambling on a retake.

In what order should I study the domains?

Start with fundamentals and the TCP/IP stack, because every later domain assumes you can read interface state and reason about layers. Then do configuration and daemon administration together, followed by routing and firewalling, which build on addressing knowledge. Finish with security, troubleshooting, and scripting, and revisit weak domains with mixed practice sets.

Which tools should I be fluent in?

Be comfortable with ip or ifconfig, ss or netstat, ping, traceroute, dig, tcpdump, iptables or nftables, and your system's service manager and logs. Know where configuration lives on your platform and how to read a routing table. Fluency means predicting output, not just recognising a command name in a list.

What are the most common failure modes?

Three recur: confusing similar tools and their flags, missing the layer at which a fault actually sits, and forgetting rule or route evaluation order. Another is neglecting logging and monitoring questions because they feel secondary. When reviewing a wrong answer, write down the exact detail that decided it, then reproduce the scenario in a lab.

Can I pass with only Linux experience?

Usually yes, but watch for BSD-flavoured assumptions. Interface naming, firewall syntax, and some daemon defaults differ between Linux and BSD, and questions may describe either. Read the wording carefully, note which platform the scenario implies, and avoid answering from habit when the question describes a different Unix variant.

Does this page or its author have official affiliation with the exam?

No. This is independent study material written for candidates preparing for the credential, and it is not endorsed by, affiliated with, or produced by the certifying organisation. Always treat the certifying body's own documentation as authoritative for objectives, logistics, and policy, and use third-party material such as this only as supplementary practice.