mirror of
https://github.com/esphome/esphome-docs.git
synced 2024-10-31 08:31:29 +01:00
b6d9eb7317
Co-authored-by: H. Árkosi Róbert <robreg@zsurob.hu>
244 lines
9.0 KiB
ReStructuredText
244 lines
9.0 KiB
ReStructuredText
Getting Started with the ESPHome Command Line
|
||
=============================================
|
||
|
||
.. seo::
|
||
:description: Getting Started guide for installing ESPHome using the command line and creating a basic configuration.
|
||
:image: console.svg
|
||
|
||
ESPHome is the perfect solution for creating custom firmwares for
|
||
your ESP8266/ESP32 boards. In this guide we’ll go through how to set up a
|
||
basic “node” in a few simple steps.
|
||
|
||
Installation
|
||
------------
|
||
|
||
See :doc:`installing_esphome`.
|
||
|
||
If you're familiar with Docker, you can use that instead!
|
||
Note that on macOS Docker `can not pass USB devices through <https://github.com/moby/hyperkit/issues/149>`__.
|
||
You will not be able to flash ESP devices through USB on Mac, all other features will work. Flashing with web dashboard is still possible.
|
||
|
||
Our image supports AMD64, ARM and ARM64 (AARCH64), and can be downloaded with:
|
||
|
||
.. code-block:: bash
|
||
|
||
docker pull ghcr.io/esphome/esphome
|
||
|
||
If you want to use `docker-compose` instead, here's a sample file:
|
||
|
||
.. code-block:: yaml
|
||
|
||
version: '3'
|
||
services:
|
||
esphome:
|
||
container_name: esphome
|
||
image: ghcr.io/esphome/esphome
|
||
volumes:
|
||
- /path/to/esphome/config:/config
|
||
- /etc/localtime:/etc/localtime:ro
|
||
restart: always
|
||
privileged: true
|
||
network_mode: host
|
||
environment:
|
||
- USERNAME=test
|
||
- PASSWORD=ChangeMe
|
||
|
||
.. note::
|
||
|
||
If you are using NFS share to back your container's config volume, you may
|
||
need to mount the volume with the `nolock` option, otherwise platformio may
|
||
freeze on container startup as per `platformIO-core Issue 3089 <https://github.com/platformio/platformio-core/issues/3089>`__
|
||
|
||
The project provides multiple docker tags; please pick the one that suits you
|
||
better:
|
||
|
||
- ``latest`` and ``stable`` point to the latest stable release available. It's
|
||
not recommended to automatically update the container based on those tags
|
||
because of the possible breaking changes between releases.
|
||
- Release-tracking tag ``YEAR.MONTH`` (e.g. ``2022.8``) points to the latest
|
||
stable patch release available within the required version. There should
|
||
never be a breaking change when upgrading the containers based on tags like
|
||
that.
|
||
- ``beta`` points to the latest released beta version, and to the latest stable
|
||
release when there is no fresh beta release.
|
||
- ``dev`` is the bleeding edge release; built daily based on the latest changes
|
||
in the ``dev`` branch.
|
||
|
||
|
||
Connecting the ESP Device
|
||
-------------------------
|
||
|
||
Follow the instructions in :doc:`physical_device_connection` to connect to your
|
||
ESP device.
|
||
|
||
.. note::
|
||
|
||
The most difficult part of setting up a new ESPHome device is the initial
|
||
installation. Installation requires that your ESP device is connected with
|
||
a cable to a computer. Later updates can be installed wirelessly.
|
||
|
||
Creating a Project
|
||
------------------
|
||
|
||
Now let’s setup a configuration file. Fortunately, ESPHome has a
|
||
friendly setup wizard that will guide you through creating your first
|
||
configuration file. For example, if you want to create a configuration
|
||
file called ``livingroom.yaml``:
|
||
|
||
.. code-block:: bash
|
||
|
||
esphome wizard livingroom.yaml
|
||
# On Docker:
|
||
docker run --rm -v "${PWD}":/config -it ghcr.io/esphome/esphome wizard livingroom.yaml
|
||
|
||
At the end of this step, you will have your first YAML configuration
|
||
file ready. It doesn't do much yet and only makes your device connect to
|
||
the WiFi network, but still it’s a first step.
|
||
|
||
Adding some features
|
||
--------------------
|
||
|
||
So now you should have a file called ``livingroom.yaml`` (or similar).
|
||
Go open that file in an editor of your choice and let’s add a :doc:`simple
|
||
GPIO switch </components/switch/gpio>` to our app.
|
||
|
||
.. code-block:: yaml
|
||
|
||
switch:
|
||
- platform: gpio
|
||
name: "Living Room Dehumidifier"
|
||
pin: 5
|
||
|
||
The configuration format should hopefully immediately seem similar to
|
||
you. ESPHome has tried to keep it as close to Home Assistant’s
|
||
``configuration.yaml`` schema as possible. In the above example, we’re
|
||
simply adding a switch that’s called “Living Room Dehumidifier” (could control
|
||
anything really, for example lights) and is connected to pin ``GPIO5``.
|
||
The nice thing about ESPHome is that it will automatically also try
|
||
to translate pin numbers for you based on the board. For example in the
|
||
above configuration, if using a NodeMCU board, you could have just as
|
||
well set ``D1`` as the ``pin:`` option.
|
||
|
||
First uploading
|
||
---------------
|
||
|
||
Now you can go ahead and add some more components. Once you feel like
|
||
you have something you want to upload to your ESP board, simply plug in
|
||
the device via USB and type the following command (replacing
|
||
``livingroom.yaml`` with your configuration file):
|
||
|
||
.. code-block:: bash
|
||
|
||
esphome run livingroom.yaml
|
||
|
||
You should see ESPHome validating the configuration and telling you
|
||
about potential problems. Then ESPHome will proceed to compile and
|
||
upload the custom firmware. You will also see that ESPHome created a
|
||
new folder with the name of your node. This is a new PlatformIO project
|
||
that you can modify afterwards and play around with.
|
||
|
||
If you are running docker on Linux you can add ``--device=/dev/ttyUSB0``
|
||
to your docker command to map a local USB device. Docker on Mac will not be able to access host USB devices.
|
||
|
||
.. code-block:: bash
|
||
|
||
docker run --rm --privileged -v "${PWD}":/config --device=/dev/ttyUSB0 -it ghcr.io/esphome/esphome run livingroom.yaml
|
||
|
||
Now when you go to the Home Assistant "Integrations" screen (under "Configuration" panel), you
|
||
should see the ESPHome device show up in the discovered section (although this can take up to 5 minutes).
|
||
Alternatively, you can manually add the device by clicking "CONFIGURE" on the ESPHome integration
|
||
and entering "<NODE_NAME>.local" as the host.
|
||
|
||
.. figure:: /components/switch/images/gpio-ui.png
|
||
:align: center
|
||
|
||
After the first upload, you will probably never need to use the USB
|
||
cable again, as all features of ESPHome are enabled remotely as well.
|
||
No more opening hidden boxes stowed in places hard to reach. Yay!
|
||
|
||
Adding A Binary Sensor
|
||
----------------------
|
||
|
||
Next, we’re going to add a very simple binary sensor that periodically
|
||
checks if a particular GPIO pin is pulled high or low - the :doc:`GPIO Binary
|
||
Sensor </components/binary_sensor/gpio>`.
|
||
|
||
.. code-block:: yaml
|
||
|
||
binary_sensor:
|
||
- platform: gpio
|
||
name: "Living Room Window"
|
||
pin:
|
||
number: 16
|
||
inverted: true
|
||
mode:
|
||
input: true
|
||
pullup: true
|
||
|
||
This is an advanced feature of ESPHome. Almost all pins can
|
||
optionally have a more complicated configuration schema with options for
|
||
inversion and pinMode - the :ref:`Pin Schema <config-pin_schema>`.
|
||
|
||
This time when uploading, you don’t need to have the device plugged in
|
||
through USB again. The upload will magically happen “over the air”.
|
||
Using ESPHome directly, this is the same as from a USB cable, but
|
||
for docker you need to supply an additional parameter:
|
||
|
||
.. code-block:: bash
|
||
|
||
esphome livingroom.yaml run
|
||
# On docker
|
||
docker run --rm -v "${PWD}":/config -it ghcr.io/esphome/esphome run livingroom.yaml
|
||
|
||
.. figure:: /components/binary_sensor/images/gpio-ui.png
|
||
|
||
Where To Go Next
|
||
----------------
|
||
|
||
Great 🎉! You’ve now successfully set up your first ESPHome project
|
||
and uploaded your first ESPHome custom firmware to your node. You’ve
|
||
also learned how to enable some basic components via the configuration
|
||
file.
|
||
|
||
So now is a great time to go take a look at the :doc:`Components Index </index>`.
|
||
Hopefully you’ll find all sensors/outputs/etc. you’ll need in there. If you’re having any problems or
|
||
want new features, please either create a new issue on the `GitHub issue
|
||
tracker <https://github.com/esphome/issues/issues>`__ or find us on the
|
||
`Discord chat <https://discord.gg/KhAMKrd>`__ (also make sure to read the :doc:`FAQ <faq>`).
|
||
|
||
Bonus: ESPHome dashboard
|
||
------------------------
|
||
|
||
ESPHome features a dashboard that you can use to easily manage your nodes
|
||
from a nice web interface. It was primarily designed for
|
||
:doc:`the Home Assistant add-on <getting_started_hassio>`, but also works with a simple command.
|
||
|
||
To start the ESPHome dashboard, simply start ESPHome with the following command
|
||
(with ``config/`` pointing to a directory where you want to store your configurations)
|
||
|
||
.. code-block:: bash
|
||
|
||
# Install dashboard dependencies
|
||
pip install tornado esptool
|
||
esphome dashboard config/
|
||
|
||
# On Docker, host networking mode is required for online status indicators
|
||
docker run --rm --net=host -v "${PWD}":/config -it ghcr.io/esphome/esphome
|
||
|
||
# On Docker with MacOS, the host networking option doesn't work as expected. An
|
||
# alternative is to use the following command if you are a MacOS user.
|
||
docker run --rm -p 6052:6052 -e ESPHOME_DASHBOARD_USE_PING=true -v "${PWD}":/config -it ghcr.io/esphome/esphome
|
||
|
||
|
||
After that, you will be able to access the dashboard through ``localhost:6052``.
|
||
|
||
.. figure:: images/dashboard_states.png
|
||
|
||
See Also
|
||
--------
|
||
|
||
- :doc:`cli`
|
||
- :doc:`ESPHome index </index>`
|
||
- :doc:`getting_started_hassio`
|
||
- :ghedit:`Edit`
|