This is a boiler plate project for exposing python functionality as webservices and containerizing them with docker

Installing Docker

Windows

1. Go to https://www.docker.com/products/docker-toolbox
2. Choose the appropriate OS and download docker tool box
3. Install docker tool box

OSX

1. Install homebrew http://brew.sh/
2. Install homebrew cask https://caskroom.github.io/
3. Install docker-toolbox using the following command: brew cask install docker-toolbox

Making sure the installation is right and setting up the OS (Windows Only)

1. Go to Start->All Programs->Oracle VM VirtualBox

2. If there is a virtual machine called default, right click and close it, then delete it by selecting remove (This is because the default VM that the installer creates doesn't use the right operating system)

3. Go to Start->All Programs->Docker and select Docker Quickstart Terminal

4. This would open a terminal and create a docker machine. This might take some time.

5. Once done, this would be seen. The IP address of the newly created docker machine would be displayed, make a note of it

                        ##         .
                  ## ## ##        ==
               ## ## ## ## ##    ===
           /"""""""""""""""""\___/ ===
      ~~~ {~~ ~~~~ ~~~ ~~~~ ~~~ ~ /  ===- ~~~
           \______ o           __/
             \    \         __/
              \____\_______/
   
    docker is configured to use the default machine with IP 192.168.99.100
    For help getting started, check out the docs at https://docs.docker.com
   

   Start interactive shell

6. Now go back to the Oracle VM VirtualBox and select the new default vm, right click and open the settings window. The Type and version of the box must be linux. Set Type to Linux and Version to Linux 2.6 /3.x/4.x (64 Bit)

Running the boiler plate code

Windows

1. Go to the quick start terminal

2. Go to the path where this project is downloaded/cloned (using usual cd commands)

3. When inside the boilerPlatePythonWebService folder where the Dockerfile is located, type

    docker build .
   

4. This would build a local image. Once completed, to check if the image has been created, type

    docker images
   

5. This must list the images on your local machine

   $ docker images

   REPOSITORY TAG IMAGE ID CREATED SIZE

   963f6de58dd5 9 seconds ago 541.3 MB

   c78f57260a05 2 minutes ago 537 MB

   f38ba9e4e61f 33 minutes ago 698.4 MB

6. Now to run it, type

    docker run -p 5000:5000 image_id
   

   for instance to run the first image listed above, the run command would be like docker run -p 5000:5000 963f6de58dd5 (add -d tag to make it run as a background thread)

7. Once started, open your browser and go to http://:5000. Mostly the docker ip address would be 192.168.99.100. So open http://192.168.99.100:5000

8. If it displays a welcome message, you are done!!

OSX

1. Open a terminal

2. Start the default docker machine:

    docker-machine start
   

3. Run the following to configure your shell

    eval $(docker-machine env)
   

4. Clone this repository:

    git clone --depth=1 https://github.com/verdeeco/boilerPlatePythonWebService.git <your-project-name>
   

5. cd into the project root and build the docker image with docker build .

6. Check to make sure the image was created with docker images

7. To run the image, execute docker run -p 5000:5000 image_id

8. Get the IP of your running docker host with docker-machine env

    export DOCKER_TLS_VERIFY="1"
    export DOCKER_HOST="tcp://192.168.99.100:2376"
    export DOCKER_CERT_PATH="/Users/gmuller/.docker/machine/machines/default"
    export DOCKER_MACHINE_NAME="default"
   

9. Navigate to http://192.168.99.100:5000 to see the running web service

Editing the boiler plate code

1. Add all dependency installs to the requirements.txt file
2. Edit the Docker file and look for #edit tags and follow instructions
3. For instructions on webservices go to http://flask.pocoo.org/ as the boilerplate uses flask framework