Telemetry

This program collects information about how Loud ML is used and saves it into a database.

More into details, it collects:

• count of docker pull images from Docker Hub;
• user environment of running instances of Loud ML deamon.

Because of regulation, personnal information cannot be stored in a database. This applies to e-mail for instance.

Requirements

InfluxDB listening on localhost:8086 and with a telemetry database. The database can be created with the command:

$ influx
> create database telemetry

Deployment

Sample NGINX configuration file:

# /etc/nginx/default.d/telemetry.conf
location /api {
  proxy_pass http://127.0.0.1:8080/api;
}

Data Visualization

Chronograf can be used to visualize the data from the InfluxDB database. Here is an example of a query in InfluxQL language:

SELECT count(distinct("hostid")) AS "count_hostid"
FROM "telemetry"."autogen"."telemetry"
WHERE time > :dashboardTime: GROUP BY time(:interval:) FILL(previous)