Prometheus S.M.A.R.T ctl metrics exporter

This is a simple exporter for the Prometheus metrics using smartctl. The script smartprom.py also comes with smartprom.service so that you can run this script in the background on your Linux OS via systemctl. The script will use port 9902, you can change it by changing it directly in the script. This script exports all of the data available from the smartctl.

Install

Note: You don't have to do this if you use the Docker image.

1. Copy the smartprom.service file into /etc/systemd/system folder.
2. Copy the smartprom.py file anywhere into your system.
3. Modify ExecStart= in the smartprom.service so that it points to smartprom.py in your system.
4. Run chmod +x smartprom.py
5. Install Python dependencies for the root user, example: sudo -H python3 -m pip install -r requirements.txt
6. Run systemctl enable smartprom and systemctl start smartprom
7. Your metrics will now be available at http://localhost:9902

Docker usage

No extra configuration needed, should work out of the box. The privileged: true is required in order for smartctl to be able to access drives from the host.

Docker image is here: https://hub.docker.com/r/matusnovak/prometheus-smartctl

The architectures supported by this image are: linux/386, linux/amd64, linux/arm/v6, linux/arm/v7, linux/arm64/v8, linux/ppc64le, linux/s390x

Example docker-compose.yml:

version: '3'
services:
  smartctl-exporter:
    image: matusnovak/prometheus-smartctl:latest
    container_name: smartctl-exporter
    privileged: true
    ports:
      - "9902:9902"
    restart: unless-stopped

Example docker-compose.yml with node-exporter and file export:

version: "3"
services:
    node-exporter:
        image: quay.io/prometheus/node-exporter
        restart: always
        volumes:
            - '/:/host:ro,rslave'
            - './tmp/:/tmp/'
        network_mode: "host"
        pid: "host"
        command:
            - "--path.rootfs=/host"
            - "--collector.textfile.directory=/tmp/"
    smartctl-exporter:
      image: matusnovak/prometheus-smartctl:latest
      container_name: smartctl-exporter
      privileged: true
      environment:
        - "SMARTCTL_METRICS_FILE_ENABLE=True"
      volumes:
        - ./tmp/:/metrics/
      restart: unless-stopped

Your metrics will be available at http://localhost:9902/metrics

The exported metrics looks like these:

smartprom_smart_passed{drive="/dev/sda",model_family="Seagate BarraCuda 3.5 (SMR)",model_name="ST6000DM003-2CY296",serial_number="WCT362XM",type="sat",user_capacity="6001175126016"} 1.0
smartprom_exit_code{drive="/dev/sda",model_family="Seagate BarraCuda 3.5 (SMR)",model_name="ST6000DM003-2CY296",serial_number="WCT362XM",type="sat",user_capacity="6001175126016"} 0.0
smartprom_raw_read_error_rate{drive="/dev/sda",model_family="Seagate BarraCuda 3.5 (SMR)",model_name="ST6000DM003-2CY296",serial_number="WCT362XM",type="sat",user_capacity="6001175126016"} 83.0
smartprom_raw_read_error_rate_raw{drive="/dev/sda",model_family="Seagate BarraCuda 3.5 (SMR)",model_name="ST6000DM003-2CY296",serial_number="WCT362XM",type="sat",user_capacity="6001175126016"} 2.23179896e+08
smartprom_power_on_hours{drive="/dev/sda",model_family="Seagate BarraCuda 3.5 (SMR)",model_name="ST6000DM003-2CY296",serial_number="WCT362XM",type="sat",user_capacity="6001175126016"} 73.0
smartprom_power_on_hours_raw{drive="/dev/sda",model_family="Seagate BarraCuda 3.5 (SMR)",model_name="ST6000DM003-2CY296",serial_number="WCT362XM",type="sat",user_capacity="6001175126016"} 24299.0
smartprom_airflow_temperature_cel{drive="/dev/sda",model_family="Seagate BarraCuda 3.5 (SMR)",model_name="ST6000DM003-2CY296",serial_number="WCT362XM",type="sat",user_capacity="6001175126016"} 60.0
smartprom_airflow_temperature_cel_raw{drive="/dev/sda",model_family="Seagate BarraCuda 3.5 (SMR)",model_name="ST6000DM003-2CY296",serial_number="WCT362XM",type="sat",user_capacity="6001175126016"} 40.0
...

If you are using a MegaRAID card to connect the drives, the metrics will export look like these:

smartprom_power_on_hours_raw{drive="megaraid,0",model_family="Western Digital Ultrastar He10/12",model_name="WDC WD80EMAZ-00M9AA0",serial_number="XXXXXXXX",type="sat",user_capacity="6001175126016"} 28522.0
smartprom_power_on_time_hours{drive="megaraid,1",model_family="Unknown",model_name="HGST HUH728080AL5200",serial_number="XXXXXXXX",type="scsi",user_capacity="6001175126016"} 37341.0

Configuration

All configuration is done with environment variables.

• SMARTCTL_REFRESH_INTERVAL: (Optional) The refresh interval of the metrics. A larger value reduces CPU usage. The default is 60 seconds.
• SMARTCTL_EXPORTER_PORT: (Optional) The address the exporter should listen on. The default is 9902.
• SMARTCTL_EXPORTER_ADDRESS: (Optional) The address the exporter should listen on. The default is to listen on all addresses.
• SMARTCTL_METRICS_FILE_ENABLE: (Optional) To enable metrics file, if you have a node exporter running anyway, you can simply read out this file . The default is False.
• SMARTCTL_METRICS_FILE_PATH: (Optional) the path, this must then also be specified in the docker-compose as volume. The default is /metrics/.

Grafana dashboard

There is a reference Grafana dashboard in grafana/grafana_dashboard.json.