smartmontools

Additional Notes

smartmontools is a package of utilities for monitoring storage device health using the S.M.A.R.T. (Self-Monitoring, Analysis and Reporting Technology) system. The two primary commands are smartctl (command-line query and control) and smartd (background daemon for proactive monitoring). S.M.A.R.T. is built into most modern ATA/SATA, SCSI/SAS, and NVMe drives.

S.M.A.R.T. monitors various drive attributes such as reallocated sector counts, spin-up time, temperature, error rates, and power-on hours. By tracking these attributes over time, imminent drive failures can often be predicted before data loss occurs. The daemon smartd can be configured to send email alerts or run scripts when thresholds are exceeded.

Syntax

smartctl [options] device
smartd [options]

Parameters

• device: The storage device path, e.g., /dev/sda, /dev/nvme0.

Common Options (smartctl)

• -i, --info: Display device identity information (model, serial, firmware, version).
• -H, --health: Show a brief health status (PASSED or FAILED).
• -A, --attributes: Display all S.M.A.R.T. attributes and their values.
• -a, --all: Show all S.M.A.R.T. information for the device.
• -l error: Show the error log.
• -l selftest: Show the self-test log.
• -t short: Run a short self-test in background.
• -t long: Run a long/extended self-test in background.
• -t conveyance: Run a conveyance self-test (quick check after transport).
• -c, --capabilities: Show device S.M.A.R.T. capabilities and self-test status.
• -s on/off: Enable or disable S.M.A.R.T. on the device.
• -x, --xall: Show extensive information including vendor-specific details.
• -d type: Specify the device type (ata, scsi, nvme, sat, usbprolific, etc.).

Common Options (smartd)

• -d, --debug: Run in debug mode (foreground, no daemonization).
• -i, --interval N: Set the check interval in seconds (default: 1800).
• -A prefix, --attributelog prefix: Log attribute changes to files.
• -q, --quiet: Suppress log messages during normal operation.
• -p pidfile: Specify a PID file location.
• -s, --savestates prefix: Save device states to files.

Examples

smartctl -i /dev/sda

Show identity information for /dev/sda.

smartctl -H /dev/nvme0n1

Check the health status of an NVMe drive.

smartctl -A /dev/sda

Display all S.M.A.R.T. attributes for /dev/sda.

smartctl -a /dev/sdb > smart_sdb.txt

Save all S.M.A.R.T. information for /dev/sdb to a file.

smartctl -t long /dev/sda

Start a long self-test on /dev/sda (background, poll with -l selftest).

smartctl -l error /dev/sda

Show the error log, which records past read/write/seek errors.

smartd -d -i 3600

Run smartd in debug mode with a 1-hour check interval.

Key S.M.A.R.T. Attributes

• Reallocated_Sector_Ct: Sectors that have been remapped to spare areas. Increasing values indicate a failing drive.
• Current_Pending_Sector: Sectors waiting to be remapped. A sign of instability.
• Raw_Read_Error_Rate: Frequency of read errors (varies by manufacturer interpretation).
• Spin_Retry_Count: Number of failed spin-up attempts.
• Temperature_Celsius: Drive temperature. Sustained high temperatures reduce drive lifespan.
• Power_On_Hours: Total hours the drive has been powered on.
• Wear_Leveling_Count (SSD/NVMe): Indicates remaining NAND flash lifetime for SSDs.

Practical Notes

• S.M.A.R.T. can provide early warning of mechanical failure, but not all failures are predicted (e.g., sudden electronic failure).
• A "PASSED" health status does not guarantee the drive is healthy; it only means no critical thresholds have been crossed.
• Monitor Reallocated_Sector_Ct over time; any increase warrants investigation and backup.
• Run smartctl -t long on new drives to establish a baseline.
• smartd is configured via /etc/smartd.conf.
• Enabling S.M.A.R.T. usually requires -s on for ATA drives; most modern drives have it enabled by default.
• NVMe drives use a different health information format (accessed via NVMe commands).
• For SSDs, monitor Media_Wearout_Indicator and Percent_Lifetime_Remaining attributes.