Put Ubuntu in Raspi, put Docker on it, and control GPIO with python from the container

Introduction

It has become a confusing title, but it looks like this

environment

Development PC: Mac Mojave RaspBerry Pi4 memory 4GB SD card 16GB

Write Ubuntu 19.10 to SD

https://ubuntu.com/download/raspberry-pi I downloaded the 64-bit version from and wrote it using balena Etchar.

Install docker on Ubuntu

Enter Ubuntu with ssh or something

sudo apt-get install docker-ce

You can install it with.

sudo usermod -aG docker ubuntu

You can docker without using sudo.

Make a base image

You can write a python script on the development PC and then copy it to dockerImage with Dockerfile, but it was difficult to understand, so once you make a base with Dockerfile, enter a container and create an image from that container.

Dockerfile

[Building a Docker environment for TensorFlow with Raspberry Pi](https://qiita.com/kohbis/items/53c27ce6c62a7039aaa2#raspberrypi-%E3%81%AB-docker-%E3%82%92%E3%81%88%E3 % 81% 84% E3% 82% 84) I referred to this article, almost above. I wanted to put rpi.gpio, but I didn't know how to put it, so I just wrote it.

FROM resin/rpi-raspbian:stretch

RUN echo "deb http://mirrordirector.raspbian.org/raspbian/ stretch main contrib non-free rpi firmware" > /etc/apt/sources.list

RUN apt-get update -y
RUN apt-get install -y --no-install-recommends \
        vim git less wget \
        build-essential \
        libatlas-base-dev \
        python3-pip python3-dev python3-setuptools\
        python3-scipy python3-h5py \
        libraspberrypi-bin \
    && apt-get clean \
    && rm -rf /var/lib/apt/lists/*

RUN pip3 install --upgrade pip setuptools
RUN pip3 install rpi.gpio

Then build

docker build -t rpi/rpi:0.1 .

Make sure you can do it with docker images! It should be done.

Start a container and write a python script

Start the container from the image you created earlier.

docker run --name rpi_test -ti --privileged rpi/rpi:0.1 /bin/bash

Then, it will be inside the container, so install what you want. Because it is a nano school

sudo apt-get update
sudo apt-get install nano

Go to home and make a script. I think it's better to set a working directory around here ...

cd home
nano test.py

It is a program that puts an LED on GPIO17 and makes an appropriate L flickering.

import RPi.GPIO as GPIO
import time
GPIO.setmode(GPIO.BOARD)
GPIO.setup(11, GPIO.OUT)
while True:
    GPIO.output(11, True)
    time.sleep(2)
    GPIO.output(11, False)
    time.sleep(2)

After saving the script

exit

Please log out with

docker ps -a
CONTAINER ID        IMAGE                 COMMAND                  CREATED             STATUS                     PORTS               NAMES
e700d8490be8        rpi/rpi:0.1           "/usr/bin/entry.sh /…"   11 minutes ago      Exited (0) 6 seconds ago                       rpi_test

The container is stopped. To make an image from this container

docker commit rpi_test rpi/rpi:0.2
docker images
REPOSITORY           TAG                 IMAGE ID            CREATED             SIZE
rpi/rpi              0.2                 50f7f2a406e7        2 minutes ago       622MB

You now have an image with the script saved.

Start-up

sudo docker run --privileged -it -d --name rpi -w /home rpi/rpi:0.2 python3 test.py

The container moves in the background and keeps fluttering! If you want to stop it, stop with docker stop rpi

Summary

I want to get sensor data soon