The modular structure of Home Assistant enables communication with other devices or services and expands Home Assistant with additional functions.

What is an integration in Home Assistant?

Home Assistant supports a large number of different manufacturers and integrates their devices or services via so-called integrations. The integrations ensure, among other things, that different devices can be made available uniformly in Home Assistant.
An example of an integration would be a Zigbee gateway: Home Assistant communicates with the gateway via the integration, whereby its devices - for example, connected lamps or sensors - can be made available uniformly in Home Assistant. Depending on the devices, communication does not necessarily have to take place via a gateway: Certain WLAN devices are usually addressed directly via an integration using the IP address (see: Shelly). In addition to providing devices, an integration can also be used to add certain services to Home Assistant, such as the Home Assistant Community Store, see HACS.

What are entities in Home Assistant?

Entities in Home Assistant reflect certain values, details, or properties of a device. For example, a temperature sensor may provide certain values as an entity:the temperature, humidity, air pressure, or battery status. Another example would be the properties of a lamp, this is likely to have as an entity a switch for turning it on and off and possibly another entity for adjusting the brightness or its light color.

ZigBee integration using the example of the Conbee2 - USB stick

To be able to use ZigBee devices with Home Assistant, I use the ConBee2 USB stick. Alternatively to the gateway solution of the manufacturer: Phoscon I recommend the use of ZigBee2MQTT or even better ZHA, see: HA: 3 different integrations for the ConBee 2 USB stick.

Home Assistant Community Store (HACS)

Even though Home Assistant supports many integrations, thanks to the use of the community, the range can be extended by countless additional integrations. HACS is an integration for downloading GitHub repositories via a kind of software catalog. The prerequisite for using the community integrations is the installation of the Home Assistant Community Store. The download of the Home Assistant Community Store starts in the terminal, which can be called up slightly differently depending on the Home Assistant installation.

Docker

If you run Home Assistant in Docker, you can install HACS via the container's terminal:

user@rechner:~$ docker exec -it home-assistant bash
bash-5.0# wget -q -O - https://install.hacs.xyz | bash -

HAOS

With a HAOS installation, the terminal can be called via an add-on.

Error: 502: Bad Gateway

Any complex password is still missing at this point. Simple passwords are not accepted at this point ...

Installation

The download and installation can be started with the following command:

wget -q -O - https://install.hacs.xyz | bash -

Activating HACS integration

The store is activated via: Settings, Integration

Integrations installed via a HACS repository are marked as "Custom Integration":

System Monitor

The "System Monitor" integration can be used for rudimentary monitoring of the operating system on which Home Assistant is running:

Monitor server or Linux systems

If you want to monitor the status of other Linux systems, you can use the integration "Glances". Glances collect data about CPU, memory or the free space of a hard disk and much more. As a prerequisite, Glances must first be installed on the systems to be monitored, see also: Glances - Monitor Linux Server

Shelly

An example of an integration that communicates directly with a WLAN device is the WLAN switch Shelly:

Since Home Assistant makes it easy to connect different systems, I use a WLAN switch in addition to ZigBee. Shelly is a small flush-mounted WLAN switch for controlling e.g. a power socket.

After setup, the device is automatically assigned to the default home dashboard:

Device Tracker OpenWrt

If you use the free OpenWrt router software, you can connect it to Home Assistant and use it to track your presence status.

device_tracker:
  - platform: luci
    host: 192.168.1.2
    username: root
    password: ???
    interval_seconds: 30
    consider_home: 360
    new_device_defaults:
       track_new_devices: true

For OpenWRT the package luci-mod-rpc is required.

For information on OpenWrt, see: Alternative Router Firmware: OpenWrt and no longer DD-WRT

Alternatively, the status can also be queried via a ping:

Ping to a network device

If you use OpenWrt to detect the presence of devices on the network via a can do this via the Ping (ICMP) integration:

The status of the sensor is "disconnected" or "connected" depending on the accessibility of the device

InfluxDB

To evaluate certain sensor data outside of Home Assistant, Grafana, for example, can be used for visualization. To make the data available in Grafana, certain sensor data can be transferred to an external InfluxDB. The InfluxDB runs in parallel with the Home Assistant database. I have added the following block to the configuration.yaml file, to connect Home Assistant to the v1 Influx database:

influxdb:
  include:
    entities:
       - sensor.flowmeter
       - sensor.heating_vorlauf
       - sensor.heating_ruecklauf
       - sensor.aussen_temperature
       - sensor.heating_warmwasser
       - sensor.heating_pufferspeicher
       - sensor.heating_waermetauscher
       - sensor.heating
       - sensor.heating_active_power
       - sensor.heizstab_power
  host: influxdb
  port: 8086
  database: ha
  username: ha
  password: ???
  max_retries: 3

Connect MQTT devices as sensor

After adding the MQTT integration, a new sensor can be created in the configuration.yaml file:

MQTT requires an MQTT broker server, see MQTT - Broker : Docker Setup.

Adding individual sensors in the configuration.yaml file:

sensor:
  - platform: mqtt
    state_topic: "flowmeter/flow"   
    name: "flowmeter"
    unit_of_measurement: "m³/h"

Services that use the MQTT broker:

• Zigbee2MQTT, see : Commissioning Zigbee2MQTT in Docker - step by step
• "AI-on-the-edge-device" digitizer, see : Monitor water meter, record consumption: ESP32-Cam
• smart meter, see : Read out smart meter of the network operator KAIFA MA309M