# Overview
# General structure
PRTG entities:
- Device
- Sensor
- Sensor type (lots of them pre-installed)
- Sensor channel
- Sensor schedule
- Alarm
- Alarm rule
- Alarm action
PRTG structures itself around a device map.
This can become confusing if services are monitored which are located
at different devices, e.g. "shop".
Each device can have multiple sensors.
A sensor collects data using a schedule; In most cases it's
"every minute 24/7".
PRTG features numerous sensor types. The most used ones are:
- Ping
- TCP port check
- SNMP query
Most sensors are performed by PRTG, but there's also the option to send sensor data form external resources to PRTG. The most viable one is the "webhook" sensor: It creates a cryptic URL which can be called via http/https. It expects incoming data in a specific format defined by PRTG.
Sensors contain at least one channel. Channels are used to represent
sub aspects of a sensor, e.g. the SNMP "filesystem free space" sensor
may have the channels "/", "/var" and "/opt".
Each sensor contains an alarm configuration. Alarms consist of
a rule and an action. The rule defines when to react, e.g. if the
sensor is on "error" for 60 seconds. The action defines what to do.
Actions are defined on global scope; Sensors refer to them.
Examples:
- Notify into Teams channel
- Notify by EMail
- Call a webhook
Sensors perform these tasks:
- execute the sensor code
- log the results per channel
- check the current result against the alarm rules
- perform alarm actions