# Ubuntu/Debian
sudo apt update && sudo apt install -y smartmontools

# CentOS/Rocky Linux
sudo dnf install -y smartmontools

# Verify installation
smartctl --version

# List all detected storage devices
sudo smartctl --scan
# Output example:
# /dev/sda -d scsi # /dev/sda, SCSI device
# /dev/sdb -d scsi # /dev/sdb, SCSI device
# /dev/nvme0 -d nvme # /dev/nvme0, NVMe device

# Get overall health status
sudo smartctl -H /dev/sda
# SMART overall-health self-assessment test result: PASSED

# Show all SMART information for a drive
sudo smartctl -a /dev/sda

# More detailed output including error log
sudo smartctl -x /dev/sda

# For SSDs, check wear indicators specifically:
sudo smartctl -A /dev/sda | grep -E "Wear|Media_Wearout|SSD_Life"

# Show drive identity information
sudo smartctl -i /dev/sda

# Show all attributes with current, worst, threshold values
sudo smartctl -A /dev/sda

# Critical attributes to watch:
# ID  1  Raw_Read_Error_Rate     — Hardware errors reading from disk surface
# ID  5  Reallocated_Sector_Ct   — Sectors with read errors remapped to spares
#                                   ANY non-zero value is concerning
# ID  9  Power_On_Hours          — Total operating hours
# ID 10  Spin_Retry_Count        — Motor spin-up failures (HDD only)
# ID 177 Wear_Leveling_Count     — SSD wear (lower = more worn)
# ID 187 Reported_Uncorrect      — Errors that couldn't be recovered
# ID 188 Command_Timeout         — Commands that timed out
# ID 196 Reallocated_Event_Count — Reallocation events (even one is a warning)
# ID 197 Current_Pending_Sector  — Sectors waiting to be reallocated
#                                   Non-zero = imminent data loss risk
# ID 198 Offline_Uncorrectable   — Sectors found bad during offline tests
# ID 199 UDMA_CRC_Error_Count    — Data transmission errors (cable/interface)

# Script to check critical attributes
sudo smartctl -A /dev/sda | awk '
/Reallocated_Sector_Ct|Current_Pending_Sector|Offline_Uncorrectable/ {
  if ($10 > 0) {
    print "WARNING: " $2 " = " $10 " (non-zero indicates potential failure)"
  }
}'

# Short test (1-2 minutes) — checks electronics and surface
sudo smartctl -t short /dev/sda

# Long test (several hours) — full surface scan
sudo smartctl -t long /dev/sda

# Conveyance test (for drives after transport, ~5 minutes)
sudo smartctl -t conveyance /dev/sda

# Offline test (runs in background without impacting performance)
sudo smartctl -t offline /dev/sda

# Check test status (run after starting a test)
sudo smartctl -a /dev/sda | grep -A 10 "Self-test execution status"

# View test history
sudo smartctl -l selftest /dev/sda

# Cancel a running test
sudo smartctl -X /dev/sda

Num  Test_Description    Status                  Remaining  LifeTime(hours)
# 1  Extended offline    Completed without error       00%      12345
# 2  Short offline       Completed without error       00%      12340

# Show NVMe SMART health information
sudo smartctl -a /dev/nvme0

# Key NVMe health fields to monitor:
sudo smartctl -a /dev/nvme0 | grep -E \
  "Critical Warning|Temperature:|Available Spare|Percentage Used|Data Units|Error Information"

# Critical Warning: 0x00 is healthy; non-zero indicates a problem
# Available Spare: should be above Available Spare Threshold
# Percentage Used: close to 100% means drive is near end of life
# Media and Data Integrity Errors: should be 0

# NVMe-specific: list namespaces
sudo nvme list   # requires nvme-cli: apt install nvme-cli

# NVMe SMART log via nvme-cli
sudo nvme smart-log /dev/nvme0

# Check error log
sudo nvme error-log /dev/nvme0

# Monitor NVMe temperature
sudo smartctl -a /dev/nvme0 | grep "Temperature"

# Configure smartd
sudo tee /etc/smartd.conf > /dev/null <<'EOF'
# Monitor all drives
DEVICESCAN -d auto \
  -H \                         # Check overall health
  -l error \                   # Report error log changes
  -l selftest \                # Report self-test log changes
  -f \                         # Report attribute failures
  -s (S/../.././02|L/../../6/03) \  # Short test daily at 2am, long test Saturdays at 3am
  -m [email protected] \       # Email address for alerts
  -M exec /usr/share/smartmontools/smartd-runner  # Email handler

# Monitor a specific drive with more aggressive testing
/dev/sda -d sat -a \
  -o on \                      # Enable automatic offline testing
  -S on \                      # Enable attribute auto-save
  -s (S/../.././04|L/../../7/04) \
  -m [email protected]
EOF

# Enable and start smartd
sudo systemctl enable smartd
sudo systemctl start smartd

# Check smartd is running
sudo systemctl status smartd

# View smartd logs
sudo journalctl -u smartd -n 50

# Install mail utilities for smartd email alerts
sudo apt install -y mailutils postfix   # Ubuntu
sudo dnf install -y mailx postfix       # CentOS/Rocky

# Test that email is working
echo "Test from smartmontools" | mail -s "Test Alert" [email protected]

# Configure smartd to send alerts on attribute change
# In /etc/smartd.conf, the -m flag sets the recipient:
# -m [email protected] -M once    # Send once per problem (default)
# -M daily                        # Send daily while problem persists
# -M diminishing                  # Send 1st, 2nd, 4th, 8th day...
# -M test                         # Send test email on startup

# Test the alert configuration
sudo smartd --quit -d -M test 2>&1 | head -20

# Custom alert script
sudo tee /usr/local/bin/smart-alert.sh > /dev/null <<'EOF'
#!/bin/bash
# Called by smartd when a problem is detected
# Arguments: see smartd.conf manpage for $SMARTD_* variables
echo "SMART Alert: $SMARTD_MESSAGE" | \
  mail -s "DISK HEALTH WARNING: $SMARTD_DEVICEPATH" [email protected]

# Also log to syslog
logger -t smartd "ALERT: $SMARTD_MESSAGE on $SMARTD_DEVICEPATH"
EOF

sudo chmod +x /usr/local/bin/smart-alert.sh

# Reference the script in smartd.conf:
# -m [email protected] -M exec /usr/local/bin/smart-alert.sh

# Create a health summary script for all drives
sudo tee /usr/local/bin/disk-health-check.sh > /dev/null <<'EOF'
#!/bin/bash
echo "=== Disk Health Report: $(date) ==="
echo ""

for DEV in $(smartctl --scan | awk '{print $1}'); do
  echo "--- $DEV ---"
  MODEL=$(smartctl -i $DEV 2>/dev/null | grep "Device Model\|Model Number" | awk -F: '{print $2}')
  HOURS=$(smartctl -A $DEV 2>/dev/null | grep "Power_On_Hours" | awk '{print $10}')
  HEALTH=$(smartctl -H $DEV 2>/dev/null | grep "result:" | awk '{print $NF}')
  
  echo "  Model: $MODEL"
  echo "  Hours: $HOURS"
  echo "  Health: $HEALTH"
  
  # Check for concerning attributes
  PENDING=$(smartctl -A $DEV 2>/dev/null | grep "Current_Pending_Sector" | awk '{print $10}')
  REALLOCATED=$(smartctl -A $DEV 2>/dev/null | grep "Reallocated_Sector_Ct" | awk '{print $10}')
  
  [ "$PENDING" -gt 0 ] 2>/dev/null && echo "  *** WARNING: Pending sectors: $PENDING ***"
  [ "$REALLOCATED" -gt 0 ] 2>/dev/null && echo "  *** WARNING: Reallocated sectors: $REALLOCATED ***"
  echo ""
done
EOF

sudo chmod +x /usr/local/bin/disk-health-check.sh

# Run it
sudo /usr/local/bin/disk-health-check.sh

# Schedule it with cron
echo "0 6 * * * root /usr/local/bin/disk-health-check.sh | mail -s 'Daily Disk Health' [email protected]" | \
  sudo tee /etc/cron.d/disk-health-check

# Try with explicit device type
sudo smartctl -d sat /dev/sda    # For SATA drives behind adapters
sudo smartctl -d scsi /dev/sda   # For SAS drives
sudo smartctl -d nvme /dev/nvme0 # For NVMe

# For drives behind RAID controllers, use specific driver
sudo smartctl -d megaraid,0 /dev/sda   # LSI/Avago MegaRAID slot 0
sudo smartctl -d 3ware,0 /dev/twa0     # 3Ware RAID

# Check if SMART is enabled on the drive
sudo smartctl -i /dev/sda | grep "SMART support is"

# Enable SMART if disabled
sudo smartctl -s on /dev/sda

# Test postfix is working
echo "test" | mail -s "test" [email protected]

# Check smartd logs for email errors
sudo journalctl -u smartd | grep -i "mail\|email"

# Test with smartd's test mode
sudo smartd -q onecheck -d