A monitoring stack that makes use of Grafana for visualizations.
It also contains Prometheus to collect metrics and serve them to Grafana. Grafana uses both Prometheus en the Telraam HTTP-API to collect and show data. Data transformations are limited in Grafana, because of this the timesync folder contains code that collects, transforms and exposes data and metrics in a simple flask webserver to be used by Grafana.
Use it together with the mityri mock tools to instantly observe stats and metrics of a near real setup.
Use it together with the banshee alert watcher to be sure you don't miss the prometheus alerts.
- Status overview: Status of the running services and devices. Pings devices and makes http requests to web servers. Shows a map of the stations. Tracks the time sync between our services.
- Baton battery percentage
- Baton battery last seen time
- Baton restarted status: if a baton uptime changes more then 3s between detections it is seen as restarted. This means something is going wrong and you want to check the baton for malfuctioning parts. If everything is fine you can set this restart status as fine by posting to the endpoint
/reset_rebooted/<mac>. Restarting the timesync server re-mark this baton as restarted.
- Data overview of telraam. Just a page with tables and a simple lap overview
- WIP: Lap insight page.
- See laps over time.
- Detect sudden lap duration differences
- Postgres overview: Monitors the postgres behavior using a postgres exporter for prometheus: https://github.com/prometheus-community/postgres_exporter
- docker-compose
Start the server with
$ ./start.sh <env-file>
In an environment file, you configure IPs and ports of the Stations, Telraam and ManualCount.
Examples are hosts_prod.env and host_simsalabim_kelder.env.
After starting, you can change IPs in the environment file. Stop the monitoring, run restore.sh and then start.sh again.
Go to localhost:3000 to see the grafana dashboard.
The preset credentials username is admin, password is in the environment file.