Monitor HDD Temperatures with Collectd, InfluxDB, Grafana – (CentOS 7.x x64)

Introduction

I like to monitor things. I like to have data. As much of it as I can. I'm the software version of a hoarder, never wanting to delete anything and to have a long, visible, track record of history for usage to be further analyzed later on.

This configuration is about monitoring the hard drive temperatures that are connected to your physical server/device.

Prerequisites

You should..

• Have Grafana installed and configured
• Have InfluxDB installed and configured
• Have collectd installed and configured
• Epel-release installed

  yum install epel-release -y

Install "hddtemp" from Epel Repository

This is the package you will need the epel-release repository for.

1. Install "hddtemp" and enable it to run as a daemon at boot time.

   yum install hddtemp -y --enablerepo=epel
   systemctl enable hddtemp.service

2. Start the service

   systemctl start hddtemp.service

Edit /etc/collectd.conf to use "hddtemp"

3. Edit the configuration for collectd to use the hddtemp plugin

   LoadPlugin hddtemp

4. Add to the collectd configuration to include the port/IP the daemon is listening on:

   <Plugin hddtemp>
           Host "127.0.0.1"
           Port 7634
   </Plugin>

   Note: If you're monitor HDD temperatures on a different physical host, then you will want that host to be running HDDTemp so that collectd can connect to it's IP address:port to pull measurements.

5. Restart collectd

   systemctl restart collectd.service

Conclusion

After some time, you should start having hard drive temperature measurements accumulate within Grafana/InfluxDB for monitoring.