AppDaemon is a loosely coupled, multithreaded, sandboxed python execution environment for writing automation apps for Home Assistant home automation software. As of release 2,0,0 it also provides a configurable dashboard (HADashboard) suitable for wall mounted tablets.
Installation is either by pip3 or Docker. There is also an official hass.io build.
Follow the instructions in DOCKER_TURORIAL.md
Before running AppDaemon you will need to install the package:
$ sudo pip3 install appdaemonThere is an official hass.io addon for AppDaemon maintained by [vkorn](https://community.home-assistant.io/u/vkorn/summary). The repository is [here](https://github.com/vkorn/hassio-addons/tree/master/appdaemon).
When you have appdaemon installed by either method you are ready to start working on the appdaemon.yaml file. For docker users, you will already have a skeleton to work with. For pip users, you need to create a configuration directory somewhere (e.g. /home/homeassistant/conf) and create a file in there called appdaemon.yaml.
Your initial file should look something like this:
AppDaemon:
logfile: STDOUT
errorfile: STDERR
threads: 10
cert_path: <path/to/root/CA/cert>
cert_verify: True
time_zone: <time zone>
HASS:
ha_url: <some_url>
ha_key: <some key>
# Apps
hello_world:
module: hello
class: HelloWorldha_urlis a reference to your home assistant installation and must include the correct port number and scheme (http://orhttps://as appropriate)ha_keyshould be set to your key if you have one, otherwise it can be removed.logfile(optional) is the path to where you wantAppDaemonto keep its main log. When run from the command line this is not used - log messages come out on the terminal. When running as a daemon this is where the log information will go. In the example above I created a directory specifically for AppDaemon to run from, although there is no reason you can't keep it in theappdaemondirectory of the cloned repository. Iflogfile = STDOUT, output will be sent to stdout instead of stderr when running in the foreground, if not specified, output will be sent to STDOUT.errorfile(optional) is the name of the logfile for errors - this will usually be errors during compilation and execution of the apps. Iferrorfile = STDERRerrors will be sent to stderr instead of a file, if not specified, output will be sent to STDERR.threads- the number of dedicated worker threads to create for running the apps. Note, this will bear no resembelance to the number of apps you have, the threads are re-used and only active for as long as required to tun a particular callback or initialization, leave this set to 10 unless you experience thread starvationcert_path(optional) - path to root CA cert directory - use only if you are using self signed certs.cert_verify(optional) - flag for cert verification - set toFalseto disable verification on self signed certs.time_zone(optional) - timezone for AppDaemon to use. If not specified, AppDaemon will query the timezone from Home Assistantad_key(optional) - adds the requirement for AppDaemon API calls to provide a key in the header of a request
Optionally, you can place your apps in a directory other than under the
config directory using the app_dir directive.
e.g.:
app_dir = /etc/appdaemon/appsThe #Apps section is the configuration for the Hello World program
and should be left in place for initial testing but can be removed later
if desired, as other Apps are added, App configuration is fully described in
the API doc.
To add an initial test app to match the configuration above, we need to first create an apps subdirectory under the conf directory. Then create a file in the apps directory called hello.py, and paste the followinginto it using your favorite text editor:
import appdaemon.appapi as appapi
#
# Hello World App
#
# Args:
#
class HelloWorld(appapi.AppDaemon):
def initialize(self):
self.log("Hello from AppDaemon")
self.log("You are now ready to run Apps!")With this app in place we will be able to test the App part of AppDaemon when we first run it.
Configuration of the dashboard component (HADashboard) is described separately in the Dashboard doc
There are a number of example apps under conf/examples in the git repository, and the
conf/examples.yaml file gives sample parameters for them.
Assuming you have set the config up as described in the tutotial for Docker, you should see the logs output as follows:
$ docker logs appdaemon
2016-08-22 10:08:16,575 INFO Got initial state
2016-08-22 10:08:16,576 INFO Loading Module: /export/hass/appdaemon_test/conf/apps/hello.py
2016-08-22 10:08:16,578 INFO Loading Object hello_world using class HelloWorld from module hello
2016-08-22 10:08:16,580 INFO Hello from AppDaemon
2016-08-22 10:08:16,584 INFO You are now ready to run Apps!Note that for Docker, the error and regular logs are combined.
You can run AppDaemon from the command line as follows:
$ appdaemon -c /home/homeassistant/confIf all is well, you should see something like the following:
$ appdaemon -c /home/homeassistant/conf 2016-08-22 10:08:16,575 INFO Got initial state 2016-08-22 10:08:16,576 INFO Loading Module: /home/homeassistant/conf/apps/hello.py 2016-08-22 10:08:16,578 INFO Loading Object hello_world using class HelloWorld from module hello 2016-08-22 10:08:16,580 INFO Hello from AppDaemon 2016-08-22 10:08:16,584 INFO You are now ready to run Apps!
usage: appdaemon [-h] [-c CONFIG] [-p PIDFILE] [-t TICK] [-s STARTTIME]
[-e ENDTIME] [-i INTERVAL]
[-D {DEBUG,INFO,WARNING,ERROR,CRITICAL}] [-v] [-d]
optional arguments:
-h, --help show this help message and exit
-c CONFIG, --config CONFIG
full path to config diectory
-p PIDFILE, --pidfile PIDFILE
full path to PID File
-t TICK, --tick TICK time in seconds that a tick in the schedular lasts
-s STARTTIME, --starttime STARTTIME
start time for scheduler <YYYY-MM-DD HH:MM:SS>
-e ENDTIME, --endtime ENDTIME
end time for scheduler <YYYY-MM-DD HH:MM:SS>
-i INTERVAL, --interval INTERVAL
multiplier for scheduler tick
-D {DEBUG,INFO,WARNING,ERROR,CRITICAL}, --debug {DEBUG,INFO,WARNING,ERROR,CRITICAL}
debug level
-v, --version show program's version number and exit
-d, --daemon run as a background process
-c is the path to the configuration directory. If not specified,
AppDaemon will look for a file named appdaemon.cfg first in
~/.homeassistant then in /etc/appdaemon. If the directory is not
specified and it is not found in either location, AppDaemon will raise
an exception. In addition, AppDaemon expects to find a dir named
apps immediately subordinate to the config directory.
-d and -p are used by the init file to start the process as a daemon and are not required if running from the command line.
-D can be used to increase the debug level for internal AppDaemon operations as well as apps using the logging function.
The -s, -i, -t and -s options are for the Time Travel feature and should only be used for testing. They are described in more detail in the API documentation.
AppDaemon also currently supports a legacy ini style of
configuration and it is shown here for backward compatibility. It is
recommended that you move to the YAML format using the provided tool.
When using the legacy configuration style, there are no HASS or
HADashboard sections - the associated directives all go in the
AppDaemon section.
[AppDaemon]
ha_url = <some_url>
ha_key = <some key>
logfile = STDOUT
errorfile = STDERR
threads = 10
cert_path = <path/to/root/CA/cert>
cert_verify = True
# Apps
[hello_world]
module = hello
class = HelloWorldIf you want to move from the legacy ini style of configuration to
YAML, AppDaemon is able to do this for you. Just run AppDaemon providing the configuration directory using the -c option as usual and specify the --convertcfg flag. From the command line run:
$ appdaemon -c YOUR_CONFIG_DIR --convertcfg
Converting /etc/appdaemon/appdaemon.cfg to /etc/appdaemon/appdaemon.yaml
$AppDaemon should correctly figure out where the file is to convert form your existing configuration. After conversion, the new YAML file will be used in preference to the old ini file, which can then be removed if desired.
Note: any lines in the ini file that are commented out, whether actual comments of lines that are not active, will not be converted. Note 2: Docker users will unfortunately need to perform the conversion manually.
To run AppDaemon at reboot, you can set it up to run as a systemd
service as follows.
Add Systemd Service (appdaemon@appdaemon.service)
First, create a new file using vi:
$ sudo vi /etc/systemd/system/appdaemon@appdaemon.service
Add the following, making sure to use the correct full path for your
config directory. Also make sure you edit the User to a valid user
to run AppDaemon, usually the same user as you are running Home
Assistant with is a good choice.
[Unit] Description=AppDaemon After=home-assistant@homeassistant.service [Service] Type=simple User=hass ExecStart=/usr/local/bin/appdaemon -c <full path to config directory> [Install] WantedBy=multi-user.target
- The above should work for hasbian, but if your homeassistant service
is named something different you may need to change the
After=lines to reflect the actual name.
$ sudo systemctl daemon-reload $ sudo systemctl enable appdaemon@appdaemon.service --now
Now AppDaemon should be up and running and good to go.
Since AppDaemon under the covers uses the exact same APIs as the frontend UI, you typically see it react at about the same time to a given event. Calling back to Home Assistant is also pretty fast especially if they are running on the same machine. In action, observed latency above the built in automation component is usually sub-second.
To update AppDaemon after new code has been released, just run the following command to update your copy:
$ sudo pip3 install --upgrade appdaemonIf you are using docker, refer to the steps in the tutorial.
AppDaemon runs under windows and has been tested with the official 3.5.2 release. There are a couple of caveats however:
- The
-dor--daemonizeoption is not supported owing to limitations in the Windows implementation of Python. - Some internal diagnostics are disabled. This is not user visible but may hamper troubleshooting of internal issues if any crop up
AppDaemon can be installed exactly as per the instructions for every other version using pip3.
Windows 10 now supports a full Linux bash environment that is capable of running Python. This is essentially an Ubuntu distribution and works extremely well. It is possible to run AppDaemon in exactly the same way as for Linux distributions, and none of the above Windows Caveats apply to this version. This is the reccomended way to run AppDaemon in a Windows 10 and later environment.