CompTIA A+ (220-1201 and 220-1202) - High-Yield Scenarios and Patterns¶
These are original help desk scenarios written to match the style of the objectives. Each one walks through the analysis a technician should do, then the fix, then how to prevent it. They are not vendor exam questions.
Core 1: Hardware Scenarios¶
New Drive Not Showing Up¶
Scenario: A technician installs a second SSD in a desktop for extra storage. The drive appears in UEFI, but not in File Explorer.
Analysis: - Domain: Core 1 Hardware and Troubleshooting, Core 2 OS tools - Clue: UEFI sees it, so power and data cables are fine - Cause: A new disk is not initialized and has no volume or drive letter
Response: 1. Open Disk Management (diskmgmt.msc) 2. Initialize the disk as GPT 3. Create a new simple volume, format NTFS, assign a drive letter
Prevention: - Add "initialize and format" to the SOP for drive installs
Laptop Clock Wrong Every Morning¶
Scenario: A user's older laptop shows the wrong date and time every time it is fully powered off. Web pages show certificate errors until the clock is fixed.
Analysis: - Clue: Time resets after power loss - Cause: The CMOS battery (coin cell) is dead - Side effect: Certificates look invalid when the clock is far off
Response: - Replace the CMOS battery (usually a CR2032) following the service manual - Reset date and time in UEFI, confirm Windows time syncs with NTP
Prevention: - Replace CMOS batteries during hardware refreshes of older machines
Choosing a RAID Level¶
Scenario: A small office wants a four-drive NAS for shared files. It must survive the loss of any two drives at once, and capacity matters less than safety.
Analysis: - Requirement: Two-drive fault tolerance, four drives - Options: RAID 5 survives one failure. RAID 10 survives one per mirror pair, not any two. RAID 6 survives any two
Response: - Configure RAID 6. Usable space is two drives' worth - Keep a separate backup. RAID does not protect against deletion or ransomware
Prevention: - Monitor drive health and replace failed drives quickly
Ghosting on Laser Prints¶
Scenario: Every page from a department laser printer shows a faint copy of the previous image lower down the page.
Analysis: - Symptom: Double/echo images (ghosting) - Cause: The drum is not fully cleaned or discharged between rotations, or the imaging unit is worn
Response: - Replace the toner/drum cartridge (or imaging unit) - If it continues, check the cleaning blade and apply the maintenance kit
Prevention: - Follow the maintenance kit schedule the printer reports
Core 1: Networking Scenarios¶
Users Cannot Get Online After a Power Cut¶
Scenario: After a power outage, several wired users in a small office report "no internet." ipconfig on their PCs shows addresses starting 169.254.
Analysis: - Clue: 169.254.x.x is APIPA - Cause: Clients cannot reach a DHCP server. The router or server providing DHCP may not have restarted correctly
Response: 1. Check the router or DHCP server is powered and running 2. Check the switch between users and the router 3. Run ipconfig /release then ipconfig /renew on a client 4. Confirm a valid address and test the gateway with ping
Prevention: - Put network gear on a UPS so brief outages do not reboot it
Slow and Dropping Wi-Fi¶
Scenario: Staff on the second floor complain Wi-Fi drops several times an hour. A Wi-Fi analyzer shows the office AP on 2.4 GHz channel 6 with 40 MHz width, and five neighboring networks on channels 4 to 8.
Analysis: - Cause: Channel overlap and interference on a crowded 2.4 GHz band, made worse by a wide channel
Response: - Set 2.4 GHz to 20 MHz width on channel 1 or 11 (whichever is least used) - Steer capable clients to 5 GHz or 6 GHz - Consider a second AP for coverage
Prevention: - Survey with a Wi-Fi analyzer before and after changes
Cable Run Through the Ceiling¶
Scenario: A contractor needs to run Ethernet above a drop ceiling that is used as an air return, to reach an access point 70 m from the switch. The AP has no power outlet nearby.
Analysis: - Clue: Air-handling space means plenum-rated cable is required - Distance: 70 m is within the 100 m limit for copper - Power: No outlet means PoE
Response: - Use plenum-rated Cat 6 (or Cat 6a) cable - Power the AP from a PoE switch or a PoE injector matched to the AP's power class - Terminate at a patch panel, test with a cable tester
Prevention: - Label both ends and record the run in network documentation
Core 2: Operating System Scenarios¶
Feature Missing on a New Laptop¶
Scenario: A new hire's laptop came with Windows 11 Home. IT cannot join it to the domain, and BitLocker is missing.
Analysis: - Cause: Home edition does not support domain join, Group Policy Editor, BitLocker, or hosting RDP
Response: - Upgrade the license to Windows 11 Pro (in-place, keeps apps and files) - Join the domain, then enable BitLocker and store the recovery key in the directory
Prevention: - Specify Pro or Enterprise in procurement
Mapped Drive Missing¶
Scenario: A user says their S: drive disappeared after a password change. Other users are fine.
Analysis: - Clue: One user, after a password change - Cause: Saved credentials for the mapped drive are stale, or the log-in script failed
Response: 1. Update stored credentials in Credential Manager 2. Remap with net use S: \\fileserver\shared 3. Run gpupdate /force if the drive is mapped by policy, then check gpresult /r
Prevention: - Map drives by Group Policy using the signed-in user's credentials
Linux Web Server Full Disk¶
Scenario: A small Linux server stops accepting uploads. Users see "no space left on device."
Analysis: - Tools: df -h shows which filesystem is full. du -sh /var/* finds the large directories
Response: - Find and rotate or remove old logs, clear package caches (apt clean or dnf clean all) - Confirm free space with df -h
Prevention: - Configure log rotation and disk space alerts
Core 2: Security Scenarios¶
Browser Hijack¶
Scenario: A user's browser home page changed to an unfamiliar search engine. Pop-ups appear even on trusted sites, and a new toolbar was installed with free PDF software.
Analysis: - Cause: A PUP or adware bundled with the download, installing a browser hijacker and extension
Response (follow the 10 steps): 1. Verify the symptoms 2. Quarantine the PC from the network 3. Disable System Restore (Windows Home) 4. Remove the bundled program and malicious extensions, reset the browser 5. Update anti-malware, scan in Safe Mode 6. Schedule scans, re-enable System Restore, create a restore point 7. Educate the user on trusted download sources
Prevention: - Standard user accounts, application allow lists, and browser extension policies
Files Renamed and Unreadable¶
Scenario: A user calls because their documents now have a strange extension and a text file on the desktop demands payment.
Analysis: - Cause: Ransomware - Urgency: It may be encrypting network shares right now
Response: - Disconnect the PC from the network immediately (unplug the cable, turn off Wi-Fi) - Escalate to security or management following the incident response plan - Preserve evidence, do not wipe until told to - Restore files from clean offline backups after the system is rebuilt
Prevention: - 3-2-1 backups with one offline or immutable copy - Least privilege on shares, EDR, user training
Disposing of Old Drives¶
Scenario: A clinic is replacing 20 PCs. The old HDDs and SSDs held patient records.
Analysis: - Regulation: Healthcare data, so destruction must be documented - Media: Mixed HDD and SSD. Degaussing does not work on SSDs
Response: - Use a certified third-party destruction vendor to shred all drives, or drill and shred in-house - Get a certificate of destruction listing each asset tag - Update the asset inventory
Prevention: - Include disposal in the asset life cycle policy
Securing a New SOHO Router¶
Scenario: A two-person accounting firm installs a new wireless router.
Response: - Change the default admin password and disable remote administration - Update firmware - Set WPA3 (or WPA2-AES if older devices need it) with a long passphrase - Change the default SSID, create a separate guest network - Disable UPnP and WPS, and forward only needed ports
Core 2: Operational Procedure Scenarios¶
A Change That Went Wrong¶
Scenario: A technician pushes a new printer driver to all PCs at 10:00 on a Tuesday without approval. Half the PCs can no longer print.
Analysis: - Failures: No change request, no testing, no approval, no maintenance window, no rollback plan
Response: - Roll back to the previous driver - Document the incident - Submit the change properly: sandbox test, risk analysis, CAB approval, maintenance window, rollback plan, and end-user acceptance
Prevention: - Enforce change management for every non-standard change
Restoring After a Failure¶
Scenario: Backups run a full every Sunday and an incremental every other night. The server fails on Friday morning.
Analysis: - Incremental restore needs: The last full plus every incremental since
Response: - Restore Sunday's full, then Monday, Tuesday, Wednesday, and Thursday incrementals in order
Prevention: - Consider synthetic fulls to shorten restores, and test restores regularly
Prohibited Content Found During Repair¶
Scenario: While fixing a laptop, a technician sees files that appear to be illegal content.
Response: - Stop working on the device and do not open more files - Report to management (and law enforcement as policy requires) - Preserve the device as-is and start a chain of custody record - Document what was seen and when
Frustrated Executive¶
Scenario: An executive interrupts the technician, complains the laptop has "always been useless," and demands a new one today.
Response: - Stay calm, do not argue or become defensive - Listen, then clarify with open-ended questions ("What happens when it slows down?") - Restate the issue and set clear expectations: what you can do now, and what needs approval - Offer repair or replacement options within policy, then follow up later
Scenario Analysis Framework¶
When you meet any scenario question, run through these:
- Which domain is this? It tells you which vocabulary the answer uses.
- What changed? New hardware, update, password, location, or power event.
- What is the scope? One user, one device, or everyone. Scope points to local vs shared causes.
- What is the least disruptive fix that addresses the cause?
- Is there a safety, security, or legal angle? Those override convenience every time.
Common Exam Keywords and Their Meaning¶
| Keyword | Meaning |
|---|---|
| FIRST | The initial step, usually gather information or check the obvious |
| NEXT | The step after the one described, follow the methodology order |
| BEST | Several work, pick the most complete and appropriate |
| MOST likely | The most common cause of that symptom |
| LEAST | Pick the weakest, cheapest, or least disruptive option depending on context |
Related Pages¶
- Fact Sheet - Tables behind these answers
- 05 - Hardware and Network Troubleshooting - Core 1 symptom tables
- 08 - Software Troubleshooting - Core 2 symptom tables
- Practice Questions - Test yourself