Skip to content

Core 1 Domain 5: Hardware and Network Troubleshooting (28%)

Overview

This is the heaviest Core 1 domain, at 28%. Every objective starts "Given a scenario, troubleshoot..." and lists common symptoms. The skill being tested is simple to state and hard to fake: read a symptom, name the most likely cause, and pick the right next step. This note is organized as symptom tables for that reason.

The objectives are:

  • 5.1 Troubleshoot motherboards, RAM, CPUs, and power.
  • 5.2 Troubleshoot drive and RAID issues.
  • 5.3 Troubleshoot video, projector, and display issues.
  • 5.4 Troubleshoot common mobile device issues.
  • 5.5 Troubleshoot network issues.
  • 5.6 Troubleshoot printer issues.

The Troubleshooting Methodology

In the V15 objectives, CompTIA prints the troubleshooting methodology as a "competency standard" and states that it is not a formal tested objective in 220-1201. Learn it anyway. It is the thinking every scenario question expects, it still appears in training material, and it is tested on Network+.

flowchart TB
    A["1. Identify the problem<br/>(ask the user, gather info, check recent changes, back up data)"] --> B["2. Establish a theory of probable cause<br/>(question the obvious, research knowledge base)"]
    B --> C{"3. Test the theory"}
    C -->|Confirmed| D["4. Establish a plan of action<br/>and implement the solution"]
    C -->|Not confirmed| B
    D --> E["5. Verify full system functionality<br/>and implement preventive measures"]
    E --> F["6. Document findings, actions,<br/>and outcomes"]

The six-step troubleshooting methodology from the A+ objectives document. If a theory fails, you form a new one, or escalate.

Practical rules that fall out of it:

  • Back up data before making changes when the problem involves storage or the OS.
  • Question the obvious. Is it plugged in? Is the monitor on the right input? Is Wi-Fi turned off?
  • Change one thing at a time, so you know which change fixed it.
  • Escalate when the fix is outside your authority or skill.

5.1 Motherboards, RAM, CPUs, and Power

Symptom Likely causes What to do
POST beeps Firmware reporting a hardware fault before video starts. Patterns vary by BIOS vendor Look up the beep code in the motherboard manual. Common culprits: RAM not seated, no video card, CPU fault
Proprietary crash screens (BSOD on Windows, kernel panic on macOS/Linux) Faulty RAM, drivers, overheating, failing drive Note the stop code, check Event Viewer, run memory diagnostics
Blank screen No power to monitor, wrong input, loose cable, failed GPU, bad RAM (no POST) Check power and input first, try another cable and display, listen for beeps
No power Unplugged, power strip off, failed PSU, loose 24-pin, front panel switch header Check the outlet, test the PSU with a PSU tester or multimeter, reseat connectors
Sluggish performance Overheating and throttling, not enough RAM, failing drive, malware Check temperatures and Task Manager, clean dust
Overheating Dust, failed fan, dried thermal paste, blocked vents Clean with compressed air, replace fan, reapply thermal paste
Burning smell Failing PSU or component, shorted circuit Power off immediately and unplug. Inspect. Replace the part
Random shutdown Overheating, failing PSU, bad RAM Check temperatures and PSU voltages. Event Viewer shows "unexpected shutdown" (Kernel-Power 41)
Application crashes Bad RAM, driver issue, overheating Run a memory test, update drivers
Unusual noise Failing fan bearing (grinding, whining), coil whine, loose screw Identify the source. Replace failing fans
Capacitor swelling Bulging or leaking capacitors on motherboard or PSU Replace the board or PSU. Do not repair
Inaccurate system date/time Dead CMOS battery (CR2032) Replace the coin cell battery, reset date and time in UEFI

RAM checks: reseat modules, test one stick at a time, try a different slot, run Windows Memory Diagnostic (mdsched.exe) or MemTest86.

5.2 Drives and RAID

Symptom Likely causes What to do
LED status indicators Amber or red drive LEDs on servers and NAS units show a failed or failing drive Replace the flagged drive, let the array rebuild
Grinding noises Mechanical failure in an HDD Back up immediately, replace
Clicking sounds ("click of death") Head failure in an HDD Back up if still possible, replace. Consider data recovery services
Bootable device not found Wrong boot order, failed drive, corrupt boot sector, disconnected cable, UEFI vs legacy mode mismatch Check UEFI boot order and mode, reseat cables, test the drive, repair boot records
Data loss/corruption Failing drive, bad shutdown, malware Restore from backup, run chkdsk, check S.M.A.R.T.
RAID failure One or more drives failed, controller failure Identify the failed drive, replace it, rebuild. On RAID 0 the data is gone
S.M.A.R.T. failure The drive's own health monitoring predicts failure Back up now, replace proactively
Extended read/write times Failing drive, fragmentation on HDD, nearly full disk Check health, free space, defragment HDDs (never defragment SSDs, Windows runs TRIM instead)
Low IOPS Slow drive type, degraded array rebuilding, contention Move to SSD, wait for rebuild, spread the load
Missing drives in OS New drive not initialized, no drive letter, cable, driver Open Disk Management, initialize, create a volume, assign a letter
Array missing RAID controller failed or lost its configuration, driver missing Check controller BIOS and driver. Do not initialize the disks, that destroys data
Audible alarms RAID controller warning of a degraded or failed array Check the controller utility, replace the failed drive

Key point: a degraded RAID 5 array is running without protection. One more failure loses the array. Replace the failed drive quickly, and make sure backups are current before the rebuild, because rebuilds stress the remaining drives.

5.3 Video, Projector, and Display

Symptom Likely causes What to do
Incorrect input source Display set to HDMI 2 while the cable is on HDMI 1 Change the input with the monitor or projector buttons
Physical cabling issues Loose, damaged, or wrong cable. VGA with bent pins Reseat, swap the cable
Burnt-out bulb (projector) Lamps have a rated life in hours Replace the lamp and reset the lamp timer. Let it cool first
Fuzzy image Not at native resolution, analog connection, focus (projector) Set native resolution, use a digital cable, adjust focus
Display burn-in Static image left on OLED or plasma for long periods Use screen savers and auto-off. Burn-in may be permanent
Dead pixels Individual pixels stuck black (dead) or on a color (stuck) Check warranty pixel policy. Replace the panel if bad enough
Flashing screen Loose cable, refresh rate mismatch, driver, failing backlight Reseat cable, update driver, set correct refresh rate
Incorrect color display Loose or damaged cable (missing pin), wrong color profile, failing GPU Swap cable, calibrate, test another display
Audio issues HDMI/DP audio routed to the wrong device, muted display speakers Select the correct output device in the OS
Dim image Brightness setting, power saving, failing backlight or inverter, projector lamp near end of life Check settings first, then hardware
Intermittent projector shutdown Overheating (clogged filter, blocked vents), failing lamp Clean filters, improve ventilation
Sizing issues Wrong resolution or scaling, overscan on TVs Adjust resolution, scaling, and overscan settings
Distorted image Wrong resolution, keystone on projector, damaged cable Fix resolution, adjust keystone correction

5.4 Mobile Device Issues

Symptom Likely causes What to do
Poor battery health Age and charge cycles, heat Check battery health, replace battery
Swollen battery Lithium-ion failure, overcharging, heat Stop using the device immediately. Do not charge. Replace and dispose via battery recycling
Broken screen Physical damage Replace the display assembly or digitizer
Improper charging Dirty or damaged port, bad cable or adapter, underpowered charger Clean the port with a non-conductive tool, try a known good cable and charger
Poor/no connectivity Airplane mode, weak signal, SIM or eSIM issue, antenna damage Toggle radios, reseat SIM, check carrier, forget and rejoin Wi-Fi
Liquid damage Water or spills. Liquid contact indicators turn red Power off, do not charge, dry and inspect. Often not covered by warranty
Overheating Heavy apps, charging while gaming, blocked vents, failing battery Close apps, remove case, check battery
Digitizer issues Touch layer damaged or out of calibration Recalibrate, remove screen protector, replace digitizer
Physically damaged ports Bent pins, broken connectors Replace port or board
Malware Sideloaded or malicious apps Remove the app, scan, factory reset if needed. Covered further in Core 2
Cursor drift/touch calibration Trackpad or digitizer calibration, swollen battery pushing on the trackpad Recalibrate. A swollen battery under the trackpad is a classic cause on laptops
Unable to install new applications Storage full, OS too old, MDM policy blocks it Free space, update the OS, check MDM restrictions
Stylus does not work Needs charging or pairing, wrong stylus type for the screen Charge and pair, confirm compatibility
Degraded performance Storage full, old OS, too many background apps, battery throttling Free space, update, restart

5.5 Network Issues

Symptom Likely causes What to do
Intermittent wireless connectivity Interference, weak signal, overlapping channels, roaming problems Use a Wi-Fi analyzer, change channel, move or add access points
Slow network speeds Congestion, duplex mismatch, bad cable, legacy Wi-Fi client, ISP throttling Test wired vs wireless, check link speed, replace cable
Limited connectivity APIPA address (DHCP failure), wrong gateway, wrong VLAN Run ipconfig, check DHCP, verify gateway and VLAN
Jitter Variation in packet arrival times, congestion Enable QoS for voice and video, reduce congestion
Poor VoIP quality Jitter, latency, packet loss QoS, wired connection for phones, check bandwidth
Port flapping A switch port going up and down repeatedly: bad cable, failing NIC, duplex mismatch Replace cable, test NIC, check switch logs
High latency Long distance (satellite), congestion, routing issues tracert/pathping to find the slow hop
External interference Microwaves, cordless phones, Bluetooth, neighboring networks on 2.4 GHz, EMI near motors and power lines Move to ⅚ GHz, shielded cable, reroute cables away from power
Authentication failures Wrong Wi-Fi passphrase, expired certificate, 802.1X/RADIUS problem, account locked Verify credentials and certificates, check the RADIUS server
Intermittent internet connectivity ISP issues, modem overheating, DNS failures Check modem logs, test by IP vs by name, contact ISP

A fast network isolation path

flowchart TB
    A["User reports no network"] --> B{"Link light on NIC<br/>and switch port?"}
    B -->|No| B1["Check cable, port, NIC.<br/>Try a known good cable"]
    B -->|Yes| C{"ipconfig shows a valid IP?"}
    C -->|169.254.x.x| C1["DHCP problem:<br/>check server, scope, VLAN"]
    C -->|Valid| D{"Ping default gateway?"}
    D -->|Fails| D1["Local network issue:<br/>VLAN, firewall, switch"]
    D -->|Works| E{"Ping 8.8.8.8 by IP?"}
    E -->|Fails| E1["Router, ISP, or<br/>upstream firewall"]
    E -->|Works| F{"Resolve a name<br/>(nslookup)?"}
    F -->|Fails| F1["DNS problem:<br/>check DNS servers, flush cache"]
    F -->|Works| G["Application or<br/>proxy issue"]

A bottom-up path for isolating a network fault, from physical link to DNS.

5.6 Printer Issues

Symptom Likely causes What to do
Lines down the printed pages Laser: scratched drum or dirty corona wire. Inkjet: clogged nozzles Replace drum/toner cartridge, clean. Run printhead cleaning on inkjets
Garbled print Wrong or corrupt driver, wrong printer language (PCL vs PostScript) Reinstall the correct driver, clear the queue
Paper jams Worn pickup rollers, wrong paper type or weight, humid paper, debris Clear carefully, replace rollers, use correct paper
Faded prints Low toner or ink, economy mode, low print density setting Replace cartridge, change settings
Paper not feeding Worn pickup rollers, wrong tray setting, overfilled tray Clean or replace rollers, check tray configuration
Multipage misfeed (several sheets at once) Worn separation pad, static, humid paper Fan the paper, replace separation pad
Multiple prints pending in queue Stuck job at the front of the queue, printer offline Clear the stuck job, restart the Print Spooler service
Speckling on printed pages Loose toner inside the printer, damaged drum Clean the printer with a toner vacuum, replace the cartridge
Double/echo images (ghosting) Drum not fully cleaned or discharged, fuser issue Replace drum or cartridge, check cleaning blade
Grinding noise Stripped gears, fuser failure, obstruction Inspect, clear debris, replace failed assembly
Finishing issues (staple jams, hole punch) Finisher jam, wrong paper type, empty staple cartridge, full punch waste bin Clear jam, refill staples, empty punch bin
Incorrect page orientation Driver or application setting Set orientation in the print dialog or driver defaults
Tray not recognized Tray not seated, sensor fault, not enabled in the driver Reseat, enable tray in device settings
Connectivity issues IP changed, Wi-Fi dropped, wrong port, offline Use a DHCP reservation, check network, re-add printer
Frozen print queue Stuck spooler Stop the Print Spooler service, clear C:\Windows\System32\spool\PRINTERS, restart the service

Toner not fused (smears or rubs off) is also common in questions even though it is not in the list above: the fuser is failing.

Hardware Tools for Troubleshooting

Tool Use
Multimeter Measure voltage, resistance, continuity
PSU tester Quick check of all PSU rails
POST card Shows POST codes on a display when the PC has no video
Loopback plug Test NIC ports
Cable tester and toner probe Verify and trace network cables
Wi-Fi analyzer Find channel overlap and weak signal

Exam Tips and Traps

  1. Wrong date and time after power off means the CMOS battery is dead.
  2. Swollen capacitors or battery - replace, never repair.
  3. Clicking or grinding HDD - back up first, then replace.
  4. Drive not in File Explorer - check Disk Management for an uninitialized disk.
  5. Array missing - do not initialize the drives.
  6. Ghost images - drum or cleaning. Smearing - fuser. Garbled - driver.
  7. Frozen print queue - restart the Print Spooler.
  8. 169.254 address - DHCP. Ping by IP works but not by name - DNS.
  9. Jitter and VoIP - QoS. Port flapping - cable or NIC.
  10. Projector shuts off by itself - overheating, clean the filter.
  11. Cursor drifting on a laptop trackpad - consider a swollen battery.
  12. Always check the obvious first: power, cables, input source, radios on.