NVMe and SSD SMART Monitoring Setup

The EAS Station health monitoring page displays S.M.A.R.T. data from your NVMe and SATA drives, including temperature, power-on hours, media errors, and health status.

Requirements

smartmontools package installed:

# Debian/Ubuntu
sudo apt install smartmontools

# RHEL/CentOS
sudo yum install smartmontools

sudo access for the web application user

Sudo Configuration

The web application needs permission to run smartctl as root to access raw device data.

Option 1: Sudoers file (Recommended)

Create /etc/sudoers.d/eas-station-smartctl:

# Allow EAS Station web app to run smartctl without password
# Replace 'www-data' with your web app user (could be 'eas-station', 'nginx', etc.)
www-data ALL=(ALL) NOPASSWD: /usr/sbin/smartctl

# If running as a specific user:
# eas-station ALL=(ALL) NOPASSWD: /usr/sbin/smartctl

Set correct permissions:

sudo chmod 0440 /etc/sudoers.d/eas-station-smartctl
sudo visudo -c  # Verify syntax

Option 2: Add user to disk group (Less secure)

# Add web app user to disk group
sudo usermod -a -G disk www-data

# Restart web service
sudo systemctl restart eas-station

Warning: This gives the user access to raw disk devices, which is a security risk.

Verification

Test smartctl access as the web app user:

# Switch to web app user
sudo -u www-data smartctl --scan

# Test NVMe access
sudo -u www-data sudo -n smartctl -a /dev/nvme0

Check health page:
- Visit /system_health or /health
- Look for "Storage" section with SMART data
- Should show temperature, power-on hours, etc.

Check logs for permission errors:

# Application logs
journalctl -u eas-station -f | grep -i smart

# Look for errors like:
# "smartctl permission denied for /dev/nvme0"
# "Permission denied (may require root/sudo privileges)"

What SMART Data is Displayed

For NVMe drives:

Temperature (automatically converted from Kelvin to Celsius)
Power-on hours
Power cycle count
Media errors and critical warnings
Data units read/written
Percentage used
Available spare capacity
Overall health status

For SATA SSDs/HDDs:

Temperature
Power-on hours
Reallocated sector count
Pending sectors
Overall health status

Troubleshooting

No SMART data appears

Check if smartctl is installed:
```
which smartctl
smartctl --version
```

Check if devices are detected:

sudo smartctl --scan
lsblk -o NAME,TYPE,TRAN,MODEL

Test manual access:

sudo smartctl -a /dev/nvme0  # For NVMe
sudo smartctl -a /dev/sda    # For SATA

Check web app user:

# Find which user runs the web app
ps aux | grep eas-station

# Test sudo access for that user
sudo -u <username> sudo -n smartctl --scan

"Permission denied" errors

Verify sudoers configuration (Option 1 above)
Check sudoers file syntax: sudo visudo -c
Ensure correct username in sudoers file
Restart web service after sudoers changes

"smartctl not found" errors

Install smartmontools package
Check PATH includes /usr/sbin: echo $PATH
Verify location: which smartctl (usually /usr/sbin/smartctl)

Security Notes

Sudoers approach (Option 1) is more secure - limits access to only smartctl command
disk group approach (Option 2) gives broader access to all block devices
The health monitoring code runs sudo -n which fails if password is required
No passwords are ever stored or prompted

NVMe-Specific Features

The code automatically:

Detects NVMe devices by name (nvme*) and transport type
Uses -d nvme flag for NVMe devices
Skips -n standby flag (NVMe doesn't support standby mode)
Converts temperature from Kelvin to Celsius (NVMe reports in Kelvin)
Parses NVMe-specific metrics (spare capacity, data units, etc.)

Files

SMART collection code: app_utils/system.py (lines 1276-1557)
NVMe detection: app_utils/system.py:_detect_device_type() (lines 1716-1750)
Health page route: webapp/routes_public.py:/system_health (line 650)
Health template: webapp/templates/system_health.html

Back to Documentation Index

This document is served from docs/guides/SMART_SETUP.md in the EAS Station installation.