Agento collects near-realtime metrics from running Linux systems and pushes all measurements to InfluxDB for later processing or visualization.
Grafana is the perfect companion that provides a beautiful visualization of all collected metrics.
Agento is designed to be deployed with virtually zero configuration or customisation to a single or multiple Linux hosts.
Agento is a near-realtime metric collecting agent (and optional server component) for Linux hosts. It will collect metrics from Linux hosts on regular intervals. It can collect metrics every second if the user wishes.
InfluxDB is a fantastic moders time series database.
Grafana is next-generation metric visualization.
The problem is that althou InfluxDB and Grafana are both fantastic, they doesn't do much without some kind of data source.
Agento is a modern metric collecter that will make the above combo usable for measuring the state of Linux hosts without all the fuzz.
Agento collect a huge variety of metrics out of the box. Here's a few examples:
| CPU usage | System, user, idle, iowait, … |
|---|---|
| Network traffic | Packages per second, bytes per second, errors, retransmits, … |
| Disk IO | Bytes and blocks read/written per second |
| Disk Usage | Bytes and inodes used, reserved and free |
| Load | 1, 5 and 15 minutes load average |
| Memory and swap | Bytes used, cached, buffered, free etc |
| Miscellaneous | Entropy pool, blocked processes, … |
Agento is designed to require minimal configuration on the collecting agents.
If you deploy a Agento server in your setup with a hostname like agento.example.com all clients in the same domain will report to that host without any configuration at all.
If you require more advanced settings the configration is a simple text based format described in the configuration documentation.