CompTIA Network+ Practice Test
Free CompTIA Network+ practice questions in English, Chinese, and Spanish — cabling and topologies, subnetting and routing, ports and protocols, network operations, security, and troubleshooting, every one with a full explanation.
Choose a domain
Practice questions based on the CompTIA Network+ exam domains and the published standards the exam is built on (IEEE 802.3 and 802.11, IEEE 802.1Q and 802.1X, the IANA port registry, IETF RFCs, and TIA-568 cabling). CompTIA and Network+ are marks of CompTIA; this site is not affiliated with or endorsed by CompTIA. Exam objectives are revised periodically — check the current official objectives for your exam version before testing.
About the CompTIA Network+ exam
Network+ is the certification where studying stops being memorization and starts being reasoning about a system you cannot see. A network has no moving parts to look at: when something breaks, all you get is a symptom — a machine that can reach an IP address but not a name, a link that is up but slow, a laptop that associates with the access point and then gets nothing. The exam is built around that gap. Its largest single domain is troubleshooting, and troubleshooting questions are not recall questions; they describe a symptom and ask what it proves, what to check next, or which step of the method you are actually on. The two things that most often decide a pass are the two things that reward drilling rather than reading: subnetting, which has to become arithmetic you do without deliberating, and the port-and-protocol set, which has to be recalled cold. Everything else — cabling, topologies, wireless, routing, security, operations — rests on a single mental model of which layer a given fact lives at, and candidates who build that model early find the rest of the material collapses into it. This bank is organized around that structure, with the troubleshooting pool weighted like the real exam rather than treated as an afterthought. Every question is answerable from text alone: where output matters, the output is printed in the question, because a technician's real skill is reading what a tool tells them, not recognizing a screenshot. Questions are anchored to the published standards the exam is built on — IEEE, IETF and IANA documents whose values are fixed once published — rather than to vendor defaults, product names, or whichever generation of hardware happens to be current. All questions come in English, Simplified Chinese, and Spanish with a full explanation for every answer, which matters more here than on most exams: networking vocabulary is dense with English acronyms that stay in English worldwide, so a Chinese- or Spanish-speaking candidate has to carry the concept in one language and the term in another.
How to study for the Network+
Start by building the map, not by collecting facts. The OSI model is not a list to memorize; it is a filing system, and its whole value is that every other fact on the exam has exactly one drawer it belongs in. A MAC address, a switch, a broadcast storm and a VLAN tag all live at one layer; an IP address, a router, a routing loop and a TTL live at the next; a port number, a three-way handshake and a window size at the next. Once the drawers exist, new material files itself and troubleshooting gets a method — you can ask which layer a symptom implicates and stop looking anywhere else. Candidates who skip this step end up with hundreds of disconnected facts and no way to retrieve the right one under time pressure. Spend the first sessions getting to where you can place any term on a layer instantly, and where you can say what a hub, a switch, a router, a firewall and a load balancer each look at when they make a decision.
Then drill subnetting until it stops being thinking and becomes arithmetic. This is the one topic where reading about the method is nearly worthless and repetition is nearly everything. Learn the block sizes — the eight values an octet mask can take and the size of the group each one creates — and then practice going straight from a prefix to the network address, the broadcast address, the usable range and the host count without re-deriving anything. Do them in small daily batches rather than one long session, and do them until you are bored, because boredom is the signal that the operation has moved somewhere you can reach while distracted. The payoff is larger than the subnetting questions themselves: masks and gateways are the hidden cause behind a great many scenario questions, and a candidate who computes quickly can test that hypothesis in seconds instead of avoiding it. Work IPv6 the same way, but aim for fluency in reading rather than in calculating — the abbreviation rules, telling a link-local address from a global unicast one on sight, and knowing what SLAAC does.
Learn the ports and protocols by function, not as a flat table. A list of numbers memorized in numeric order collapses under exam pressure, because the exam asks in the opposite direction: it gives you a situation and wants the protocol, or gives you a protocol and wants the transport it runs over and why. Group them the way they are actually used — the remote-access pair and why one of them is unacceptable, the file-transfer family and what distinguishes each member, the mail-sending protocol against the two mail-retrieval ones and what each does to the copy on the server, the directory protocols, the management protocols, the name and address services. For each group, hold on to the reason as well as the number: why DNS uses one transport for a normal lookup, why a protocol carrying credentials in the clear has a secured replacement, why DHCP needs a relay agent to cross a router. The reason is what survives when the number briefly deserts you, and it is also what the scenario questions are really testing.
Finally, treat the troubleshooting methodology as scored content in its own right, because it is. It is tempting to skim the ordered steps as bureaucratic filler and go back to the technical material, but the exam asks about them directly — which step an described action belongs to, what comes next, when to escalate — and the ordering is not arbitrary. Practice by narrating your reasoning on scenario questions out loud: name the symptom, say what it rules in and what it rules out, state the single thing you would check next and what each outcome would prove. Two habits carry disproportionate weight. First, change one variable at a time, so that when something improves you know what did it. Second, distinguish what a symptom proves from what it merely suggests — a successful ping to an address with a failed lookup of the name proves the path is fine and points at name resolution; an address in the automatic private range proves no server answered, not that the cable is bad. Get comfortable reading tool output in that spirit and the largest domain on the exam turns into the most predictable one.
FAQ
What does the Network+ actually test?
It is a vendor-neutral exam covering the whole surface of a working network: how the OSI and TCP/IP models organize what a device does, the physical and wireless media that carry traffic, IP addressing and subnetting, routing and switching concepts, the ports and protocols that services run on, the operational side (documentation, monitoring, availability, change management), network security, and troubleshooting. Troubleshooting is the largest domain, which tells you what the exam is really after — it wants someone who can be handed a symptom and reason toward a cause, not someone who can recite definitions. Expect a mix of straight knowledge questions, self-contained subnetting calculations, and scenario questions where you pick the next action rather than the final diagnosis.
Do I really need to subnet in my head?
You need it fast enough that it does not eat your time or your confidence. Nobody expects elegance — what you need is to look at a prefix and produce the block size, the network address, the broadcast address, the usable range and the host count without re-deriving powers of two each time. The reliable method is to learn the eight possible octet values and the block size each one implies, then practice until you go from prefix to answer in one step. It is worth over-practicing this specifically, because it is the one topic on the exam where being slow costs you twice: once on the subnetting questions themselves, and again on the scenario questions where a wrong mask is the hidden cause. The addressing pool here is deliberately heavy on self-contained calculations for exactly that reason.
Where does Network+ sit next to A+ and Security+?
They form a natural progression: A+ is the endpoint and the help desk, Network+ is everything between endpoints, and Security+ is how you defend it. Network+ is the middle one for a reason — a lot of Security+ material only makes sense once you can already picture a VLAN, a routing decision, or a TLS session on a port. If your goal is a networking or infrastructure role, Network+ is the one that changes what you can be handed at work, because the moment you can read a traceroute or reason about a duplex mismatch you stop escalating and start resolving. Taking them in order is not required, but going Network+ before Security+ tends to make the second one substantially easier.
Why are there no diagrams, screenshots, or exhibits in these questions?
Because the skill worth practicing is reading and reasoning, not picture-matching. Every question here is fully self-contained in its text: where a topology matters it is described, and where the answer depends on tool output — a ping result, a traceroute, an address configuration, a log line — that output is printed in the question itself. This is deliberately harder than recognizing a familiar screenshot, and it is closer to the job, where the information arrives as a wall of text in a terminal rather than as a labelled figure. It also means the bank works identically on a phone, in a screen reader, and in all three languages, with nothing lost to an image that was only ever captioned in English.
Why don't the questions mention specific vendors, products, or exam version codes?
Because those are the parts that go out of date, and a practice bank that teaches a stale fact is worse than one that stays quiet. Vendor defaults change with firmware, product lines get renamed, menu paths move, and exam objectives are revised on their own cycle. What does not change is the standard underneath: an IEEE designation, an IANA port number, an RFC-defined address range, a TIA-568 pinout, a syslog severity level, the arithmetic of a prefix. Those are fixed the moment they are published, so every answer here is keyed to one of them or to a durable concept. You will still see command names like ping, traceroute, dig and netstat, because those are stable and genuinely tested — but never a vendor's specific flags or output formatting. Check the current official objectives for the exam version you are booked for; this bank is built to stay correct across those revisions rather than to track them.
How should I use this bank, and what score means I'm ready?
Work one domain at a time until each is comfortable, then mix them, because the real exam does not tell you which domain a question came from and half the difficulty is recognizing that. Read the explanation on every question you get right as well as every one you miss — on this material a lucky guess and a solid answer look identical in the score. When you are consistently clearing a comfortable margin on mixed sets, and specifically when subnetting no longer slows you down and you can name the troubleshooting step a scenario is describing, you are in good shape. Treat a practice percentage as a readiness signal rather than a prediction: this bank is scored as a straightforward percentage, while the real exam uses its own scaled scoring, so the number here is not directly comparable to an official result.