Skip to content

Linux Disk Diagnostics: Tools, Interpretation, and Resolution

Goal: Diagnose disk problems on Linux using command-line tools and SMART data analysis Level: Intermediate → Advanced Language: English

Table of Contents: Essential Tools | SMART Data | Self-Tests | Kernel Messages | Filesystem Integrity | Surface Scan | Software RAID | SATA Link Speed | smartd Monitoring | Diagnostic Script | Cross-Platform Comparison | Quick Reference

A practical guide to diagnosing disk problems on Linux using command-line tools. Covers the same diagnostic methodology as the Windows Disk Diagnostics guide, using Linux-native tooling.

Essential Tools

Tool Package Purpose
smartctl smartmontools SMART data query and disk self-tests
lsblk util-linux Block device layout and mount points
journalctl systemd System journal — disk errors, boot events
dmesg util-linux Kernel ring buffer — low-level hardware messages
hdparm hdparm Disk parameters, interface speed, benchmarks
blkid util-linux Filesystem type and UUID information
mdadm mdadm Software RAID status and management
fsck e2fsprogs / xfsprogs Filesystem check and repair
badblocks e2fsprogs Surface scan for bad sectors

Install smartmontools

# Debian / Ubuntu
sudo apt install smartmontools

# RHEL / CentOS / Fedora
sudo dnf install smartmontools

# Arch
sudo pacman -S smartmontools

Step 1: Identify Block Devices

List All Disks and Partitions

lsblk -o NAME,SIZE,TYPE,FSTYPE,MOUNTPOINT,MODEL,SERIAL,ROTA,TRAN
  • ROTA — 1 for spinning disk (HDD), 0 for SSD/NVMe
  • TRAN — transport type: sata, nvme, usb, etc.

Detailed Disk Information

# All SCSI/SATA/NVMe disks
lsblk -d -o NAME,SIZE,MODEL,SERIAL,TRAN,ROTA,STATE

# More detail via hdparm (SATA only)
sudo hdparm -I /dev/sda | head -40

The hdparm -I output includes the negotiated SATA link speed — useful for detecting cable issues:

 *  Gen1 signaling speed (1.5Gb/s)
 *  Gen2 signaling speed (3.0Gb/s)
 *  Gen3 signaling speed (6.0Gb/s)

Link Speed Mismatch

If the drive supports Gen3 but only negotiates Gen1, suspect a cable or port problem.

Step 2: Query SMART Data

smartctl from the smartmontools package is the primary tool for reading SMART attributes on Linux. It provides far more detail than any GUI tool.

Quick Health Check

# Overall health assessment
sudo smartctl -H /dev/sda

# Result is either PASSED or FAILED
# FAILED means the drive's own firmware considers it end-of-life

Full SMART Attribute Dump

sudo smartctl -A /dev/sda

Example output (SATA HDD):

ID# ATTRIBUTE_NAME          FLAG     VALUE WORST THRESH TYPE      UPDATED  RAW_VALUE
  1 Raw_Read_Error_Rate     0x002f   200   200   051    Pre-fail  Always       0
  5 Reallocated_Sector_Ct   0x0033   200   200   140    Pre-fail  Always       0
  7 Seek_Error_Rate         0x002e   200   200   000    Old_age   Always       0
  9 Power_On_Hours          0x0032   082   082   000    Old_age   Always   13156
197 Current_Pending_Sector  0x0012   200   200   000    Old_age   Always       0
198 Offline_Uncorrectable   0x0010   100   253   000    Old_age   Offline      0
199 UDMA_CRC_Error_Count    0x003e   200   200   000    Old_age   Always      62

Complete Drive Information

# Everything: identity, SMART attributes, error log, self-test log
sudo smartctl -a /dev/sda

# For NVMe drives
sudo smartctl -a /dev/nvme0

Interpreting SMART Attributes

Disk Surface Health

ID Attribute Healthy Critical
5 Reallocated_Sector_Ct 0 > 0 means bad sectors exist; rising count = active degradation
196 Reallocated_Event_Count 0 Number of remap operations performed
197 Current_Pending_Sector 0 Sectors queued for reallocation test
198 Offline_Uncorrectable 0 Unreadable sectors — data has been lost

Interface Health (SATA Cable / Port)

ID Attribute Healthy Critical
199 UDMA_CRC_Error_Count 0 Any value > 0 = data transfer error. Almost always a cable, connector, or port issue — not the disk itself.

General Indicators

ID Attribute Notes
1 Raw_Read_Error_Rate Vendor-specific. WD uses a complex formula — a nonzero raw value is normal.
9 Power_On_Hours Total runtime. HDDs: 30,000–50,000 hours typical lifespan.
10 Spin_Retry_Count Failed spin-up attempts. May indicate PSU or mechanical issues.
194 Temperature_Celsius Operating temp. HDDs: keep below 45°C; SSDs: below 70°C.

The Decision Tree

SMART Analysis
│
├── smartctl -H says FAILED?
│   └── YES → Drive firmware declares end-of-life. Replace immediately.
│
├── Reallocated_Sector_Ct (5) > 0?
│   ├── YES + rising → Disk is dying. Back up and replace.
│   └── YES + stable → Old damage, currently stable. Monitor closely.
│
├── Current_Pending_Sector (197) > 0?
│   └── YES → Sectors pending reallocation. May clear after rewrite,
│             or may grow. Run self-test: smartctl -t long /dev/sdX
│
├── UDMA_CRC_Error_Count (199) > 0?
│   └── YES → Cable / port / connector problem.
│             Replace cable, try different port, check power.
│
└── All zeros but I/O errors in dmesg?
    └── Intermittent connection. Reseat cables, check PSU.

Step 3: Run SMART Self-Tests

smartctl can instruct the drive to run internal diagnostics.

# Short self-test (~2 minutes)
sudo smartctl -t short /dev/sda

# Extended self-test (can take hours on large drives)
sudo smartctl -t long /dev/sda

# Check self-test results
sudo smartctl -l selftest /dev/sda

Example result:

Num  Test_Description  Status                  Remaining  LifeTime  LBA_of_first_error
# 1  Extended offline  Completed without error       00%     13158  -
# 2  Short offline     Completed without error       00%     13156  -

Self-Test Failure

If a test shows Completed: read failure with an LBA address, that's a confirmed bad sector.

Step 4: Check Kernel Messages

Recent Disk Errors from dmesg

# Filter for disk-related messages
sudo dmesg -T | grep -iE 'ata[0-9]|sd[a-z]|error|i/o|reset|failed|exception|timeout|sector'

Key dmesg Patterns

Pattern Meaning
ata1.00: exception Emask ATA command failed — cable, port, or disk issue
ata1.00: failed command: READ FPDMA Read operation failed (NCQ)
ata1: SATA link up at 1.5 Gbps Link negotiated below capability — cable problem
ata1: SATA link down Drive disconnected or cable completely failed
ata1.00: status: { DRDY ERR } Drive reported an error
ata1.00: error: { UNC } Uncorrectable read error — bad sector
ata1.00: error: { ICRC ABRT } CRC error — SATA cable problem
sd 0:0:0:0: [sda] tag#X CRC error Data transfer corruption
I/O error, dev sda, sector XXXXX Kernel could not read/write a specific sector
Buffer I/O error on dev sda Higher-level I/O failure
EXT4-fs error: ... I/O failure Filesystem detected the I/O error
ata1.00: revalidation failed Drive re-identification failed after error
ata1: hard reset / soft reset Controller reset the link to recover from errors

ICRC Errors

The pattern error: { ICRC ABRT } in dmesg is the Linux equivalent of SMART attribute 199 (UDMA CRC Error Count). It confirms a SATA interface problem — the data arriving over the cable has CRC mismatches. Replace the cable.

Using journalctl for Historical Analysis

# All disk-related messages from the last 7 days
journalctl --since "7 days ago" -k | grep -iE 'ata[0-9]|sd[a-z]|error|i/o|reset|sector'

# Boot-related messages
journalctl --list-boots

# Messages from a specific boot (e.g., the failed one)
journalctl -b -1 -k | grep -iE 'ata|disk|error|i/o'

# Count ATA errors per device
journalctl --since "7 days ago" -k | grep -c 'ata1'
journalctl --since "7 days ago" -k | grep -c 'ata2'

Step 5: Check Filesystem Integrity

ext4

# Check (read-only, can run on mounted filesystem)
sudo e2fsck -n /dev/sda1

# Repair (filesystem must be unmounted!)
sudo umount /dev/sda1
sudo e2fsck -f -y /dev/sda1

XFS

# Check
sudo xfs_repair -n /dev/sda1

# Repair (filesystem must be unmounted!)
sudo umount /dev/sda1
sudo xfs_repair /dev/sda1

Btrfs

# Check (can run on mounted filesystem in read-only mode)
sudo btrfs check /dev/sda1

# Scrub (online verification — preferred method)
sudo btrfs scrub start /mountpoint
sudo btrfs scrub status /mountpoint

Never Run fsck on a Mounted Filesystem

Running e2fsck or xfs_repair on a mounted filesystem can cause catastrophic data corruption. Always unmount first. For the root filesystem, boot from a live USB.

Step 6: Surface Scan with badblocks

For a thorough sector-by-sector scan when SMART data is inconclusive:

# Read-only test (safe, non-destructive)
sudo badblocks -sv /dev/sda

# Read-write test (DESTROYS ALL DATA — use only on empty disks)
sudo badblocks -wsv /dev/sda
  • -s shows progress
  • -v verbose output
  • -w write mode — writes patterns, reads back, compares

Performance Note

This takes a long time on large drives. Use SMART self-tests first; reserve badblocks for cases where SMART data is unavailable or suspect.

Step 7: Check Software RAID (mdadm)

If the system uses Linux software RAID:

# RAID array status
cat /proc/mdstat

# Detailed status for a specific array
sudo mdadm --detail /dev/md0

# Check for failed/degraded members
sudo mdadm --detail /dev/md0 | grep -E 'State|Failed|Active|Spare'

Key States

State Meaning
active Array is healthy
degraded One or more members failed — data is at risk
rebuilding Array is reconstructing after a disk replacement
failed Array has lost redundancy or is non-functional 
# Check negotiated vs. capable link speed
sudo hdparm -I /dev/sda | grep -A5 "SATA Version"

# Or via sysfs
cat /sys/class/ata_link/link*/sata_spd
cat /sys/class/ata_link/link*/sata_spd_limit

A mismatch between the negotiated speed and the drive's capability is a strong indicator of cable problems:

Drive Capability Negotiated Interpretation
6.0 Gb/s 6.0 Gb/s Normal
6.0 Gb/s 3.0 Gb/s Cable or port limiting speed
6.0 Gb/s 1.5 Gb/s Bad cable — fallback to minimum
Any Link down Cable disconnected or dead

Automated Monitoring with smartd

The smartd daemon from smartmontools continuously monitors SMART data and sends alerts.

Configuration

sudo vim /etc/smartd.conf

Basic configuration — monitor all drives, email on problems:

# Monitor all drives, check every 12 hours, email on issues
DEVICESCAN -a -o on -S on -n standby,q -s (S/../.././02|L/../../6/03) -W 4,45,55 -m admin@example.com

Flags explained:

Flag Meaning
-a Monitor all SMART attributes
-o on Enable automatic offline tests
-S on Enable automatic attribute autosave
-n standby,q Don't wake sleeping drives
-s (S/../.././02\|L/../../6/03) Short test daily at 2am, long test Saturdays at 3am
-W 4,45,55 Warn at 45°C, critical at 55°C, report if temp rises 4°C between checks
-m email Send alerts to this address

Enable and Start

sudo systemctl enable smartd
sudo systemctl start smartd

# Check status
sudo systemctl status smartd

The Complete Diagnostic Script

Save as disk_diag.sh and run with sudo:

disk_diag.sh — Click to expand 
#!/bin/bash
# Linux Disk Diagnostics Script
# Usage: sudo ./disk_diag.sh | tee disk_report.txt

set -euo pipefail

echo "============================================"
echo "  DISK DIAGNOSTICS REPORT"
echo "  $(date)"
echo "============================================"

# 1. Block Device Layout
echo ""
echo "--- BLOCK DEVICES ---"
lsblk -o NAME,SIZE,TYPE,FSTYPE,MOUNTPOINT,MODEL,SERIAL,ROTA,TRAN

# 2. SMART Health for Each Disk
echo ""
echo "--- SMART HEALTH STATUS ---"
for disk in /dev/sd? /dev/nvme?n1; do
    [ -b "$disk" ] || continue
    echo ""
    echo "  Disk: $disk"
    model=$(smartctl -i "$disk" 2>/dev/null | grep "Device Model\|Model Number" | head -1 | sed 's/.*: *//')
    serial=$(smartctl -i "$disk" 2>/dev/null | grep "Serial Number" | head -1 | sed 's/.*: *//')
    echo "  Model:  $model"
    echo "  Serial: $serial"

    health=$(smartctl -H "$disk" 2>/dev/null | grep -i "result\|status" | head -1)
    echo "  Health: $health"

    # Key SMART attributes (SATA)
    smartctl -A "$disk" 2>/dev/null | awk '
        $1 == 5   { printf "    Reallocated Sectors:  %s\n", $NF }
        $1 == 196 { printf "    Reallocation Events: %s\n", $NF }
        $1 == 197 { printf "    Current Pending:     %s\n", $NF }
        $1 == 198 { printf "    Offline Uncorrect:   %s\n", $NF }
        $1 == 199 { printf "    CRC Error Count:     %s%s\n", $NF, ($NF > 0 ? " <<<< CABLE ISSUE!" : "") }
        $1 == 9   { printf "    Power-On Hours:      %s\n", $NF }
        $1 == 194 { printf "    Temperature (C):     %s\n", $NF }
    '

    # NVMe attributes
    smartctl -A "$disk" 2>/dev/null | awk '
        /Percentage Used/    { printf "    Percentage Used:     %s\n", $NF }
        /Media.*Integrity/   { printf "    Media Errors:        %s\n", $NF }
        /Power On Hours/     { printf "    Power-On Hours:      %s\n", $NF }
        /Temperature:/       { printf "    Temperature:         %s\n", $NF }
    '
done

# 3. SATA Link Speeds
echo ""
echo "--- SATA LINK SPEEDS ---"
for link in /sys/class/ata_link/link*; do
    [ -d "$link" ] || continue
    spd=$(cat "$link/sata_spd" 2>/dev/null || echo "N/A")
    limit=$(cat "$link/sata_spd_limit" 2>/dev/null || echo "N/A")
    name=$(basename "$link")
    echo "  $name: negotiated=$spd limit=$limit"
done

# 4. Kernel Disk Errors
echo ""
echo "--- KERNEL DISK ERRORS (recent) ---"
dmesg -T 2>/dev/null | grep -iE 'ata[0-9].*error|ata[0-9].*exception|ata[0-9].*reset|sd[a-z].*error|i/o error|ICRC|CRC|bad sector|medium error' | tail -30 || echo "  No disk errors in kernel ring buffer."

# 5. Journal Disk Errors (last 7 days)
echo ""
echo "--- JOURNALCTL DISK ERRORS (last 7 days) ---"
error_count=$(journalctl --since "7 days ago" -k 2>/dev/null | grep -ciE 'ata.*error|i/o error|reset|ICRC' || true)
echo "  Total disk-related error messages: $error_count"

if [ "$error_count" -gt 0 ]; then
    echo ""
    journalctl --since "7 days ago" -k 2>/dev/null | grep -iE 'ata.*error|ata.*exception|i/o error|ICRC|reset' | tail -20
fi

# 6. Boot History
echo ""
echo "--- RECENT BOOTS ---"
journalctl --list-boots 2>/dev/null | tail -10

# 7. Software RAID Status
echo ""
echo "--- SOFTWARE RAID STATUS ---"
if [ -f /proc/mdstat ]; then
    cat /proc/mdstat
else
    echo "  No software RAID configured."
fi

# 8. Filesystem Mount Health
echo ""
echo "--- MOUNTED FILESYSTEMS ---"
df -hT | grep -v tmpfs | grep -v devtmpfs

echo ""
echo "--- DONE ---"
echo "For full SMART details on a specific disk: sudo smartctl -a /dev/sdX"

Windows vs. Linux: Tool Comparison

Task Windows Linux
List disks Get-PhysicalDisk lsblk -d
Disk-to-mount mapping Get-Disk + Get-Partition lsblk
SMART health check CrystalDiskInfo smartctl -H /dev/sdX
SMART attributes CrystalDiskInfo (GUI) smartctl -A /dev/sdX
SMART self-test CrystalDiskInfo smartctl -t long /dev/sdX
Disk errors (log) Get-WinEvent (Event Log) journalctl -k, dmesg
Boot history Event IDs 41, 6008 journalctl --list-boots
Filesystem check chkdsk X: /f /r e2fsck, xfs_repair
Surface scan chkdsk X: /r badblocks -sv
SATA link speed CrystalDiskInfo Transfer Mode hdparm -I, sysfs
Software RAID Storage Spaces (Get-StoragePool) mdadm --detail, /proc/mdstat
Continuous monitoring — smartd daemon
SATA cable diagnosis SMART ID 199 (CRC Error Count) SMART ID 199 + dmesg ICRC errors

Quick Reference: One-Liners

# Am I about to lose data?
sudo smartctl -H /dev/sda

# Is my SATA cable bad?
sudo smartctl -A /dev/sda | grep CRC

# What happened during the last crash?
journalctl -b -1 -k | grep -i error

# How many disk errors this week?
journalctl --since "7 days ago" -k | grep -c 'I/O error'

# What speed is my SATA link running at?
sudo hdparm -I /dev/sda | grep -A3 "SATA Version"

# Is my RAID healthy?
cat /proc/mdstat

# Start a long SMART self-test
sudo smartctl -t long /dev/sda

# Watch self-test progress
sudo smartctl -l selftest /dev/sda