This is a Python script that will continuously run the official Speedtest CLI application by Ookla, takes input from environment variables, formats data and writes it to an InfluxDB v2 database.
This script will allow you to measure your internet connections speed and consistency over time. It uses env variables as configuration. It's as easy to use as telling your Docker server a 1 line command and you'll be set. Using Grafana you can start exploring this data easily.
Breadlysm built a grafana dashboard for this data that can be found at https://grafana.com/grafana/dashboards/13053. Additionally, other contributors have modified this dash and included a JSON file of those modifications. Use `GrafanaDash-SpeedTests.json` to import that dash into Grafana.
There are some added features to allow some additional details that Ookla provides as tags on your data. Some examples are your current ISP, the interface being used, the server who hosted the test. Overtime, you could identify if some serers are performing better than others.
## Differences to InfluxDB v1 program:
* Slightly different environment variables to set up InfluxDB connection
* Speeds will be returned in bytes without transforming into MBit/s, just like the speedtest does
* The measure duplication of `speeds>bandwith_up/down` and `download>bandwith` and `upload>bandwith` has been removed, leaving only the `speeds` fields. Same thing with `ping` and `speeds>latency/jitter` duplication.
- NAMESPACE is used to collect data from multiple instances of the container into one database and select which you wish to view in Grafana. i.e. Breadlysm has one monitoring their Starlink, the other their TELUS connection.
Be aware that this script will automatically accept the license and GDPR statement so that it can run non-interactively. Make sure you agree with them before running.
This script looks to have been originally written by https://github.com/aidengilmartin/speedtest-to-influxdb/blob/master/main.py and I forked it from https://github.com/breadlysm/speedtest-to-influxdb. They did the hard work, I've continued to modify it though to fit my needs.