Getting Started
===============

Datas are stored locally and can be sent to an influxdb database an visualized into grafana.

These tools don't cover  `influxdb <https://docs.influxdata.com/influxdb/>`_
and `grafana <http://docs.grafana.org/>`_ installation, see dedicated websites for tutorial on how to do this.


Preparation
-----------
* Prepare a raspberry pi with a `RASPBIAN LITE image <https://www.raspberrypi.org/downloads/raspbian/>`_ and an
  arduino board.
* Make sure the pi and arduino are turned-off
* Insert the ardas shield in the arduino board
* Connect the arduino board to the raspberry pi with a USB cable (no external power cable for the arduino board)
* Power up the raspberry pi and wait for the raspberry pi to boot
* Log on the raspberry pi (using ssh)

.. warning:: Make sure you don't have a random cron task activated

For visualization(optional)
---------------------------

Log on you influxdb server and create a new database::

    create database DATABASE_NAME

Create a user for this database and grant him permissions on this database::

    create user USERNAME with password 'MYPASSWORD'
    grant all on "DATABASE_NAME" to "USERNAME"

Add the database as data source on grafana.

Install Git
-----------
::

    sudo apt install git

Download the code
-----------------

    To deploy ardas-tools including ardas::

        rm -rf * .git .gitignore && git init && git remote add origin https://github.com/UMONS-GFA/ardas-tools.git && git pull origin master

Install the required packages
-----------------------------
::

    ./raspbian_upgrade.sh


Initialization and setup
------------------------
* For production


    To initiate stable branch of ardas::

        bash init_ardas.sh

    To update stable branch::

        bash update_ardas.sh

    These scripts can take quite a long time to retrieve the python dependencies, espacially when using a rapsberry pi one.
    See the :ref:`options` option for an alternative...

* For developers

    To initiate development branch of ardas ::

        bash init_ardas.sh --dev

    To update development branch ::

        bash update_ardas.sh --dev

    Follow the instructions to set the approriate sensors and acquisition settings.


Configure the PYTHONPATH
------------------------
::

    nano ~/.bashrc

add at the end::

    export PYTHONPATH=/home/pi/ardas

Refresh your profile::

    source ~/.bashrc

Generate the sensors
--------------------

You can use the *uncalibrated_sensors_generator.py* script to generate 4 uncalibrated sensors::

    python3 /home/pi/ardas/ardas/uncalibrated_sensors_generator.py

Configure the ardas
-------------------
* Set the default arduino config, default netid will be 255::

        bash reset_arduino_EEPROM.sh

* Upload the sketch in the arduino::

        bash upgrade_sketch.sh


Define your settings
--------------------

Make a copy of the *settings_example.py* and name it *settings.py*::

    cp /home/pi/ardas/ardas/settings_example.py /home/pi/ardas/ardas/settings.py

Configure this file for your needs.


Add a cron task
---------------
::

    crontab -e


and add::

    PYTHONPATH=/home/pi/ardas

    # m h  dom mon dow   command
    @reboot /usr/bin/python3 /home/pi/ardas/ardas/raspardas.py > /home/pi/ardas/cronlog.log 2>&1 &



Start logging
-------------
    To reboot the raspberry.py and start logging::

        bash reboot.sh

    Don't forget to type a message to explain in which circumstances (and eventually by whom) the system is rebooted. It show in the logs when the system will restart.

.. _options:

Options
-------

* Update code without automatically rebooting(only for update_ardas script)::

        --noreboot

* Prevent creating a new virtual environment and installing all dependencies

    Installing dependencies could be slow on a raspberry pi, espacially a . If you plan to install several identical loggers,
    it could be advisable to create a disk image with the right dependencies installed in the virtual environment.
    The following option could then be used to prevent the system from creating a new virtual environment and retrieving dependencies::

       --nopip


Debug
-----

If there is a communication problem with the Arduino, you can try to debug with picocom::

    apt install picocom

Make sure the python cron task is not running::

    ps -aux | grep python3

Launch picocom::

    picocom -b 57600 /dev/ttyACM0

Use Ctr+a then Ctrl+c to enable echo.

To quit, use Ctrl+x

Call the ardas with his netid::

    -255
Set naïve data mode::

    #ND

You can now change the default config with ZR command::

    #ZR station netId integrationPeriod nbInst sensor1 sensor2 sensor3 sensor4 code

    Ex: #ZR 1111 222 3333 4 0001 0002 0003 0004 31

Set raspardas data mode::

    #RD

Change your settings accordingly