CompTIA A+ Practice Questions: 25 Exam-Style Examples With Answers and Explanations (220-1201 and 220-1202, 2026)

Most free CompTIA A+ practice questions online still target 220-1101 and 220-1102. Those exams are gone. The questions below are written to the current V15 objectives, Core 1 (220-1201) and Core 2 (220-1202), and they are weighted roughly the way the real exams are weighted, so the mix you get here reflects the mix you will face.

There are 25 questions: 13 from Core 1 and 12 from Core 2. Every one has the answer and the reason it is the answer, because a practice question you got right for the wrong reason is worth nothing on exam day.

The exams you are practising for

Core 1 splits five ways: Hardware and Network Troubleshooting 28%, Hardware 25%, Networking 23%, Mobile Devices 13%, Virtualisation and Cloud Computing 11%.

Core 2 splits four ways: Operating Systems 28%, Security 28%, Software Troubleshooting 23%, Operational Procedures 21%.

Exam Tip: Core 2 has the higher pass mark (700 against 675) and no domain below 21%. There is nowhere to hide on Core 2, whereas on Core 1 you can afford to be weaker on Virtualisation and Cloud Computing because it is only 11% of the exam.

Both exams are capped at 90 questions rather than fixed at 90, and performance-based questions sit at the front. Answering every multiple-choice question first and returning to the PBQs is the single most useful piece of timing advice for this exam.

How to use this practice set

Give yourself 30 minutes for all 25 and do not look at the answers until you have finished. Write down your reasoning for anything you are unsure of. When you check, compare your reasoning to the explanation, not just your letter to the answer letter. That comparison is where the learning happens.

Core 1 (220-1201) Practice Questions

Question 1

A user reports that their laptop screen is visible only when tilted at a sharp angle and looks very dim otherwise. Which component has most likely failed?

A. LCD panel B. Inverter or backlight C. Video cable D. Integrated GPU

Answer: B. A dim image that is still faintly visible means the panel is receiving and rendering the signal correctly, so the display itself is fine. What has failed is the light behind it. On older laptops this was the inverter driving a CCFL backlight; on modern panels it is the LED backlight circuit. A failed video cable or GPU produces artefacts, lines or no image at all, not a correctly formed but unlit one.

Question 2

A technician needs to connect a modern smartphone to a projector that has only an HDMI input. The phone has a single USB-C port. What is the correct solution?

A. A passive USB-C to HDMI cable, provided the port supports DisplayPort Alt Mode B. A USB-C to USB-A adapter followed by a USB to HDMI converter C. A Lightning to HDMI adapter D. Bluetooth screen mirroring to the projector

Answer: A. Video over USB-C depends on DisplayPort Alternate Mode, which reassigns some of the port's high-speed lanes to carry a native DisplayPort signal. If the phone supports it, a passive adapter works. If it does not, no passive adapter will help, which is why checking the port specification comes before buying the cable.

Question 3

Which cellular technology standard introduced support for network slicing and sub-6 GHz plus millimetre wave bands?

A. LTE B. 4G LTE-Advanced C. 5G D. HSPA+

Answer: C. Network slicing, which partitions one physical network into isolated virtual networks with different performance guarantees, is a 5G feature. The sub-6 GHz and mmWave band split is also specific to 5G, where sub-6 GHz gives range and mmWave gives throughput over very short distances.

Question 4

A small office needs to connect two buildings 400 metres apart with a 10 Gbps link. Which cable type is appropriate?

A. Cat 6 UTP B. Cat 6a UTP C. Single-mode fibre D. RG-6 coaxial

Answer: C. Copper twisted pair is limited to 100 metres regardless of category, so both Cat 6 and Cat 6a are out at 400 metres. Cat 6a would carry 10 Gbps but only to 100 metres. Single-mode fibre handles the distance and the bandwidth, and it also removes the earthing and lightning risk that comes with running copper between separate buildings.

Question 5

A technician is configuring a wireless access point and must choose a channel in the 2.4 GHz band that will not overlap with neighbouring networks on channels 1 and 11. Which channel should they use?

A. 3 B. 6 C. 9 D. 13

Answer: B. In the 2.4 GHz band, only channels 1, 6 and 11 are non-overlapping in most regions, because each 20 MHz channel is spaced 5 MHz apart and so bleeds into its neighbours. With 1 and 11 taken, 6 is the only choice that avoids overlap. Channels 3 and 9 would interfere with both existing networks.

Question 6

Which port number does IMAP use when secured with TLS?

A. 110 B. 143 C. 993 D. 995

Answer: C. IMAP over TLS is 993. Plain IMAP is 143, plain POP3 is 110, and POP3 over TLS is 995. The pair that catches people out is 993 and 995, so tie them to the protocol name rather than trying to remember the numbers in isolation.

Question 7

A workstation used for video editing must run four displays and a high-throughput capture card. Which motherboard specification matters most?

A. The number and generation of PCIe lanes B. The number of SATA ports C. The chipset's maximum RAM capacity D. The form factor

Answer: A. Both the GPU driving the displays and the capture card compete for PCIe lanes, and the number of lanes a platform provides is fixed by the CPU and chipset. If there are not enough lanes, slots drop to a lower width and throughput falls. RAM capacity matters for editing, but it is not what limits two bandwidth-hungry expansion cards sharing one board.

Question 8

A printer is producing pages with a repeating smudge that appears at the same interval down every page. What should the technician check first?

A. The fuser temperature B. The imaging drum C. The paper tray alignment D. The network print queue

Answer: B. A defect repeating at a fixed interval points to a rotating component, and the interval corresponds to that component's circumference. The imaging drum is the usual culprit. A failing fuser produces toner that smears when rubbed or does not adhere at all, which is a different symptom entirely.

Question 9

Which RAID level provides both striping and mirroring, and survives the loss of one drive in each mirrored pair?

A. RAID 0 B. RAID 5 C. RAID 6 D. RAID 10

Answer: D. RAID 10 mirrors pairs of drives and then stripes across those mirrors, so it tolerates one failure per pair. RAID 0 has no redundancy at all. RAID 5 survives one drive failure across the whole array and RAID 6 survives two, but neither mirrors.

Question 10

An organisation wants to run a workload in the cloud but requires that the underlying physical hardware is not shared with any other customer. Which model meets the requirement?

A. Public cloud with a dedicated host B. Multitenant public cloud C. Community cloud D. Serverless functions

Answer: A. A dedicated host allocates physical hardware to a single customer while keeping the workload inside the public cloud provider. Multitenant public cloud is the opposite by definition. A community cloud is shared between several organisations with common requirements, so the hardware is still shared.

Question 11

A desktop powers on, fans spin, but there is no video output and no POST beeps. The technician has already confirmed the monitor works on another machine. What is the most efficient next step?

A. Replace the power supply B. Reseat the RAM and clear CMOS C. Reinstall the operating system D. Replace the motherboard

Answer: B. Fans spinning confirms the power supply is delivering at least some power, so replacing it is not the first move. No POST at all points at RAM, CPU seating or firmware configuration, and reseating memory plus clearing CMOS costs minutes and resolves a large share of these cases. Replacing the motherboard is the expensive last resort, not the first test.

Question 12

Users on one floor report intermittent network drops. The switch shows a high count of CRC errors on the affected ports. What is the most likely cause?

A. IP address conflicts B. A duplex mismatch or damaged cabling C. DNS server failure D. An exhausted DHCP scope

Answer: B. CRC errors mean frames are arriving corrupted, which is a physical layer or duplex problem, not an addressing problem. Duplex mismatches and damaged or excessively long cable runs are the two classic causes. An IP conflict, DNS failure or exhausted DHCP scope would all produce connection failures without corrupting frames in transit.

Question 13

A laptop's battery is visibly swollen. What should the technician do?

A. Continue using the device until a replacement arrives B. Discharge the battery fully, then refit it C. Stop using the device, isolate it and dispose of the battery through approved handling D. Puncture the battery to release the gas

Answer: C. A swollen lithium-ion battery is a fire risk and must be taken out of service immediately. It goes to approved disposal, never into general waste, and it is never punctured or compressed. This is a safety question, and on safety questions the exam always rewards the cautious answer.

Core 2 (220-1202) Practice Questions

Question 14

Which Windows edition feature allows a device to be joined to an Active Directory domain?

A. Windows Home B. Windows Pro C. Windows Home with a Microsoft account D. Any edition with the correct licence key

Answer: B. Domain join is a Pro, Enterprise and Education feature. Windows Home cannot join a domain regardless of licensing, which is why upgrading the edition is the answer whenever a scenario describes a Home machine that needs to join a corporate domain.

Question 15

A technician needs to check and repair the integrity of Windows system files on a machine that boots but behaves erratically. Which command should they run first?

A. chkdsk /f B. sfc /scannow C. diskpart D. gpupdate /force

Answer: B. System File Checker verifies protected system files against known-good copies and repairs those that fail. chkdsk addresses file system and disk structure problems, which is a different fault. If sfc reports that it cannot repair a file, DISM is the follow-up.

Question 16

Which Linux command displays the currently mounted file systems and their available space in human-readable units?

A. du -h B. df -h C. ls -lh D. mount -a

Answer: B. df -h reports free space per mounted file system. du -h reports the size of directories and files, which is a related but different question. mount -a mounts everything listed in fstab rather than reporting anything.

Question 17

An organisation wants users to authenticate with something they know and something they have. Which pairing satisfies this?

A. A password and a security question B. A password and a hardware token C. A fingerprint and a facial scan D. A PIN and a password

Answer: B. A password is something you know and a hardware token is something you have, so the pairing covers two distinct factors. A password with a security question is two instances of something you know. A fingerprint and a facial scan are both something you are. Two of the same factor is not multifactor authentication.

Question 18

Which attack involves an attacker positioning themselves between two communicating parties to intercept or alter traffic?

A. On-path attack B. Denial of service C. SQL injection D. Zero-day exploit

Answer: A. The on-path attack, previously called man-in-the-middle, describes an attacker who sits in the communication path and can read or modify traffic in transit. Denial of service aims to make a resource unavailable rather than to intercept it, and SQL injection targets a database through unsanitised input.

Question 19

A company is disposing of hard drives that held regulated customer data and must be able to prove the data is unrecoverable. Which method is most appropriate?

A. Quick format B. Standard format C. Physical shredding with a certificate of destruction D. Deleting the partitions

Answer: C. Only physical destruction with documentation gives you proof, and proof is what a regulated disposal requires. Quick formats, standard formats and partition deletion all leave data recoverable to some degree. When a question mentions regulated data and evidence of destruction, the answer is destruction plus a certificate.

Question 20

Which Windows feature encrypts an entire volume and can bind the key to the system's TPM?

A. EFS B. BitLocker C. Windows Defender Firewall D. Credential Guard

Answer: B. BitLocker performs full volume encryption and can seal its key to the TPM so the drive only unlocks on the original hardware in an expected boot state. EFS encrypts individual files and folders rather than whole volumes, and it does not use the TPM in that way.

Question 21

A user's browser redirects to unwanted search pages and shows pop-ups even when closed. Which step comes first in the CompTIA malware removal process after identifying the symptoms?

A. Educate the user B. Quarantine the infected system C. Remediate the system D. Enable System Restore

Answer: B. The documented order is investigate and verify symptoms, quarantine the system, disable System Restore, remediate, schedule scans and updates, re-enable System Restore and create a restore point, then educate the user. Quarantine comes second because containment has to happen before cleaning, and System Restore is disabled rather than enabled so the malware is not preserved in a restore point.

Question 22

An application crashes on launch on one machine but runs on identical machines. Where should the technician look first?

A. Event Viewer application logs B. Task Manager startup tab C. Device Manager D. The registry editor

Answer: A. Event Viewer records the specific fault module and exception code for an application crash, which turns guesswork into a targeted fix. Task Manager and Device Manager tell you nothing about why a process terminated. Editing the registry before you know the cause risks creating a second problem.

Question 23

A mobile device is running hot, draining its battery quickly and using far more data than usual. What is the most likely explanation?

A. A failing battery B. A malicious or misbehaving application C. A weak cellular signal D. An outdated operating system

Answer: B. Heat, battery drain and unexpected data use together point at something running constantly in the background and transmitting. A failing battery drains but does not increase data use. A weak signal raises power draw but not data volume. The three symptoms in combination are what identify the cause.

Question 24

Which document defines the agreed uptime and response times between a service provider and a customer?

A. Acceptable use policy B. Service level agreement C. Non-disclosure agreement D. Statement of work

Answer: B. The service level agreement sets measurable commitments such as uptime percentage and response time, along with what happens when they are missed. A statement of work defines deliverables and scope, and an acceptable use policy governs how staff may use company systems.

Question 25

A change to a production system has been approved and scheduled. Which element of the change request describes what will be done if the change fails?

A. Scope of the change B. Risk analysis C. Rollback plan D. Purpose of the change

Answer: C. The rollback plan, sometimes called the backout plan, sets out how the system is returned to its previous working state. Risk analysis assesses what might go wrong beforehand, but it does not tell the engineer at 2am what to actually do about it.

Scoring yourself

These 25 questions are not scaled the way the real exam is, so treat the result as a signal rather than a prediction.

The useful number is not the total. It is which domains your wrong answers fall into. Four misses spread evenly across eight domains is a very different problem from four misses all in Hardware and Network Troubleshooting, which is 28% of Core 1.

Where candidates actually lose marks

Performance-based questions are the most common reason a well-prepared candidate runs out of time. They appear first, they take several minutes each, and candidates who work through them in order arrive at question 40 with twenty minutes left. Flag them, clear the multiple choice, then come back.

The second reason is treating Core 2 as the easy one. It carries the higher pass mark at 700, and Operating Systems and Security are 28% each. Candidates who spend eight weeks on hardware and two on Core 2 fail Core 2.

The third is memorising port numbers and command syntax in isolation. The exam asks you to apply them to a scenario. Knowing that 993 is IMAP over TLS is only useful if you recognise the scenario where a mail client fails to connect securely.

If you want a week-by-week structure rather than a question set, our 8-week CompTIA A+ study plan maps the domains to a schedule. If you have not decided between A+ and the entry-level alternative, CompTIA Tech+ versus A+ covers that choice.

Frequently Asked Questions

Is the A+ exam difficult?

A+ is broad rather than deep, and that breadth is what makes it difficult. Two exams cover nine domains spanning hardware, networking, operating systems, security and process, so there is a lot of ground to hold in memory at once. Candidates with hands-on IT experience generally find it manageable; candidates studying purely from books tend to struggle with the troubleshooting scenarios, which are 28% of Core 1 and 23% of Core 2.

How do you pass the A+ certification exam?

You need 675 out of 900 on Core 1 and 700 out of 900 on Core 2, taken as two separate exams within three years of each other. The approach that works is domain-weighted study, so more time on Troubleshooting, Hardware and Networking for Core 1 and on Operating Systems and Security for Core 2, combined with timed practice that includes performance-based questions. Hands-on time with real hardware and a Windows or Linux virtual machine is worth more than extra reading.

Is CompTIA A+ still worth it?

A+ remains the standard entry requirement for help desk and desktop support roles, and it is the qualifying credential for a large number of public sector and contractor job listings. It is not worth it if you already have two years of professional IT support experience, because employers will weigh the experience more heavily. It is worth it if you are changing careers and need something on the CV that hiring filters recognise.

Is Network+ or A+ harder?

Network+ is harder in depth, A+ is harder in breadth. Network+ goes considerably further into subnetting, routing protocols and network troubleshooting, and it expects genuine comfort with binary maths. A+ covers more ground but at a shallower level. Most candidates take A+ first for that reason, though it is not a formal prerequisite.

How many questions are on the CompTIA A+ exam?

Each Core exam has a maximum of 90 questions in 90 minutes, mixing multiple choice with performance-based simulations. The number is a cap rather than a fixed count, so some candidates see fewer. Since both exams are required, you are answering up to 180 questions across the full certification.

Ready to Start Practising?

Twenty-five questions tells you where you stand. It does not tell you whether you can hold that standard across 90 questions in 90 minutes with performance-based simulations eating your clock, and that is the thing that decides the result.

CertCrush has full-length A+ practice exams built to the current 220-1201 and 220-1202 objectives, with domain-weighted question banks and an explanation behind every answer. You can see exactly which domains are costing you marks and fix them before you pay $274 to find out in the test centre.

Start practising CompTIA A+ with CertCrush, or take a look at the CompTIA A+ Core 1 and Core 2 course first.