Skip to content

A small web server to send data from Ecowitt devices to an MQTT Broker

License

Notifications You must be signed in to change notification settings

mgorven/ecowitt2mqtt

 
 

Repository files navigation

🔘 ecowitt2mqtt: Send Ecowitt device data to an MQTT broker

PyPi Docker Hub Version License Maintainability Say Thanks

ecowitt2mqtt is a small CLI/web server that allows Ecowitt device data to be sent to an MQTT broker.

Installation

pip install ecowitt2mqtt

Python Versions

ecowitt2mqtt is currently supported on:

  • Python 3.6
  • Python 3.7
  • Python 3.8

Quick Start

Note that this README assumes:

  • you have access to an MQTT broker
  • you have already paired your Ecowitt device with the WS View Android/iOS app from Ecowitt.

First, install ecowitt2mqtt via pip:

$ pip install ecowitt2mqtt

Then, shift over to the WS View app on your Android/iOS device. While viewing your device in the app, select Weather Services:

Select Weather Services

Press Next until you reach the Customized screen:

The Customized screen in the WS View app

Fill out the form with these values and tap Save:

  • Protocol Type Same As: Ecowitt
  • Server IP / Hostname: the IP address/hostname of the device running ecowitt2mqtt
  • Path: /data/report (note that unlike the default in the WS View App, there shouldn't be a trailing slash)
  • Port: 8080 (the default port on which ecowitt2mqtt is served)
  • Upload Interval: 60 (change this to alter the frequency with which data is published)

Then, on the machine where you installed ecowitt2mqtt, run it:

$ ecowitt2mqtt \
    --mqtt-broker=192.168.1.101 \
    --mqtt-username=user \
    --mqtt-password=password \
    --mqtt-topic=ecowitt2mqtt/device_1

Within the Upload Interval, data should begin to appear in the MQTT broker.

Advanced Usage

Command Line Interface

The ecowitt2mqtt executable contains several configurable parameters:

usage: ecowitt2mqtt [-h] --mqtt-broker MQTT_BROKER --mqtt-topic MQTT_TOPIC [--mqtt-port MQTT_PORT]
                    [--mqtt-username MQTT_USERNAME] [--mqtt-password MQTT_PASSWORD]
                    [--endpoint ENDPOINT] [--port PORT] [-l LOG_LEVEL]

Send data from Ecowitt devices to an MQTT broker

optional arguments:
  -h, --help            show this help message and exit
  -l LOG_LEVEL, --log-level LOG_LEVEL
                        The logging level (default: INFO)
  --mqtt-broker MQTT_BROKER
                        The hostname or IP address of the MQTT broker
  --mqtt-port MQTT_PORT
                        The port of the MQTT broker (default: 1883)
  --mqtt-username MQTT_USERNAME
                        The username to use with the MQTT broker (default: None)
  --mqtt-password MQTT_PASSWORD
                        The password to use with the MQTT broker (default: None)
  --mqtt-topic MQTT_TOPIC
                        The MQTT topic to publish the device's data to (default: ecowitt2mqtt/<ID>)
  --hass-discovery      Publish data in the Home Assistant MQTT Discovery format
  --hass-discovery-prefix HASS_DISCOVERY_PREFIX
                        The Home Assistant discovery prefix to use (default: homeassistant)
  --endpoint ENDPOINT   The relative endpoint/path to serve the web app on (default: /data/report)
  --port PORT           The port to serve the web app on (default: 8080)
  --raw-data            Return raw data (don't attempt to translate any values)
  --input-unit-system INPUT_UNIT_SYSTEM
                        The input unit system used by the device (default: imperial)
  --output-unit-system OUTPUT_UNIT_SYSTEM
                        The unit system to use in output (default: imperial)

Unit Systems

ecowitt2mqtt allows you to specify both the input and output unit systems for a device. This is fairly self-explanatory, but take care to use an --input-unit-system that is consistent with what your device provides (otherwise, your data will be way off).

Raw Data

In some cases, it may be preferable to prevent ecowitt2mqtt from doing any data translation (converting values to a new unit system, changing binary values – such as might be used by a battery – into "friendly" values, etc.). Passing the --raw-data flag will accomplish this: data will flow directly from the Ecowitt device to the MQTT broker as-is.

Note that the --raw-data flag supersedes any that might cause data translation (such as --input-unit-system or --output-unit-system).

Home Assistant MQTT Discovery

Home Assistant users can quickly add entities from an Ecowitt device by using MQTT Discovery.

Once Home Assistant is configured to accept MQTT Discovery, ecowitt2mqtt simply needs the --hass-discovery flag:

$ ecowitt2mqtt \
    --mqtt-broker=192.168.1.101 \
    --mqtt-username=user \
    --mqtt-password=password \
    --hass-discovery

Note that if both --hass-discovery and --mqtt-topic are provided, --hass-discovery will win out.

Running in the Background

ecowitt2mqtt doesn't, itself, provide any sort of daemonization mechanism. The suggested route is to use a different application.

supervisord

An example supervisord configuration file might look like this:

[supervisord]
nodaemon=true
loglevel=info
user=root

[program:ecowitt2mqtt]
command=ecowitt2mqtt --mqtt-broker=192.168.1.101 --mqtt-username=user --mqtt-password=password
stdout_logfile=/dev/stdout
stdout_logfile_maxbytes=0
redirect_stderr=true

systemd

An example systemd service file in /etc/systemd/system might look like this:

[Unit]
Description=ECOWITT2MQTT daemon
After=network.target
 
[Service]
Type=notify
ExecStart=ecowitt2mqtt --mqtt-broker=192.168.1.101 --mqtt-username=user --mqtt-password=password
ExecReload=kill -HUP $MAINPID
KillMode=process
Restart=on-failure
RestartSec=5s
 
[Install]
WantedBy=multi-user.target

To enable the service:

$ systemctl enable ecowitt2mqtt

Docker

The library is available via a Docker image (bachya/ecowitt2mqtt). It is configured by a handful of environment variables that correspond to the command line parameters listed above:

  • LOG_LEVEL: the log level to use (default: INFO)
  • MQTT_BROKER: the hostname or IP address of the MQTT broker
  • MQTT_PORT: the port of the MQTT broker (default: 1883)
  • MQTT_PASSWORD: the password to use with the MQTT broker (default: None)
  • MQTT_USERNAME: the password to use with the MQTT broker (default: None)
  • MQTT_TOPIC: the MQTT topic to publish the device's data to
  • HASS_DISCOVERY: whether to use Home Assistant MQTT Discovery (default: false)
  • HASS_DISCOVERY_PREFIX: the topic prefix to use for Home Assistant MQTT Discovery (default: homeassistant)
  • ENDPOINT: the relative endpoint/path to serve the web app on (default: /data/report)
  • PORT: the port to serve the web app on (default: 8080)
  • INPUT_UNIT_SYSTEM: the input unit system to use (imperial or metric) (default: imperial)
  • OUTPUT_UNIT_SYSTEM: the output unit system to use (imperial or metric) (default: imperial)

Running the image is straightforward:

docker run -it \
    -e MQTT_BROKER=192.168.1.101 \
    -e MQTT_USERNAME=user \
    -e MQTT_PASSWORD=password \
    -p 8080:8080 \
    bachya/ecowitt2mqtt:latest

Note the value of the -p flag: you must expose the port defined by the PORT environment variable. In the example above, the default port (8080) is used and is exposed via the same port on the host.

docker-compose users can find an example configuration file at docker-compose.dev.yml. Note that this is intended to be a dev environment for quickly testing the repo itself; in production, you should refer to one of the Docker Hub images.

Contributing

  1. Check for open features/bugs or initiate a discussion on one.
  2. Fork the repository.
  3. (optional, but highly recommended) Create a virtual environment: python3 -m venv .venv
  4. (optional, but highly recommended) Enter the virtual environment: source ./.venv/bin/activate
  5. Install the dev environment: script/setup
  6. Code your new feature or bug fix.
  7. Update README.md with any new documentation.
  8. Add yourself to AUTHORS.md.
  9. Submit a pull request!

About

A small web server to send data from Ecowitt devices to an MQTT Broker

Resources

License

Stars

Watchers

Forks

Packages

No packages published

Languages

  • Python 90.7%
  • Shell 7.8%
  • Dockerfile 1.5%