Docker is the most used software for creating containers. In one of the articles I have explained what Docker is and how we can containerise a simple NodeJS application. In this article we will go deeper to understand some of the advanced features.
Docker Compose
With docker compose you can configure and start multiple containers with a single yaml file. This is really helpful if you are working on a technology stack with multiple technologies. As an example, if you are working on a project that requires MySQL database, python for AI/ML, NodeJS for real-time processing & .NET for serving API’s. It would be cumbersome to setup such an environment for each team member. Docker alleviates this issue with the help of compose.
Understanding Docker Compose
docker compose is a yaml file in which we can configure different types of services. Then with a single command all containers built and fired up. There are mainly 3 steps involved
Generate a Dockerfile for each project.
Setup services in docker-compose.yml file.
Fire up the containers.
We are now going to implement a solution for the above problem statement.
Prerequisites
One might expect to have all the technologies installed to run the above discussed tech stack of MySQL, Python, NodeJS, .NET & PHP. Rather all you need is a docker engine running. The latest versions of docker comes with docker compose installed. For now no other installation is required.
Roadmap
Before we start here is just a brief of what we are going to do. We will be tackling each technology one by one. For each technology we will create a sample application (except MySQL) and create a Dockerfile for each. Then we will point these Dockerfile in our docker compose yaml file. Lastly, we will configure docker compose so that each application does what it’s supposed to do.
Before we get started create a folder name super-app. Secondly, create a docker-compose.yml file. In this file we will configure all our application’s. So let’s get started!
Configurations
MySQL
Set the below content in your docker-compose.yml file.
version: '3.4'
services:
super-app-db:
image: mysql:8.0.28
environment:
MYSQL_DATABASE: 'super-app'
MYSQL_ROOT_PASSWORD: '$SuperApp1'
ports:
- '3306:3306'
expose:
- '3306' Under the services section we will list down all the types of applications to be configured. To start with we configure a super-app-db service which pulls a docker image of MySQL with version 8.0.28. Next, we instruct the container to create a database name super-app with root as default user and it’s password set to $SuperApp1. Lastly, since the default port for MySQL is 3306 we are mapping it to the containers’ port 3306 and exposing that port for access.
Once the above file is created run the below command to get your docker image created and run it as a container.
docker compose up MySQL image will be pulled and then the docker will spin up a container to run this image. MySQL server can be verified by connecting it via a MySQL client. If not, worry not, we will see below how to connect to it via one of our applications. As long as the container is not deleted the tables will be persisted.
Let’s configure our next application NodeJS.
NodeJS
We will create a very simple express node application. In order to do so create a folder named node inside our super-app folder. Add the below 3 files namely server.js, package.json & Dockerfile in the node folder.
const server = require("express")();
server.listen(3000, async () => { });
server.get("/super-app", async (_, response) => {
response.json({ "super": "app" });
}); {
"name": "super-app-node",
"dependencies": {
"express": "^4.17.1"
}
} # Download the slim version of node
FROM node:17-slim
# Set the work directory to app folder.
# We will be copying our code here
WORKDIR /node
#Copy package.json file in the node folder inside container
COPY package.json .
# Install the dependencies in the container
RUN npm install
# Copy the rest of the code in the container
COPY . .
# Run the node server with server.js file
CMD ["node", "server.js"]
# Expose the service over PORT 3000
EXPOSE 3000 Here we are creating a node application that returns json when we hit localhost:3000/super-app in the browser. Now, we are not directly going to run the project from this folder.
Instead, go back to your super-app folder and append the below lines to your docker-compose.yml file.
super-app-node:
build: ./node
ports:
- "3000:3000" We are simply mentioning to create a service named super-app-node. We are also mapping container port to the host port 3000. Finally run the below command to run your two containers i.e. MySQL & NodeJS.
docker compose up Now if you hit localhost:3000/super-app you will see a response {“super”:”app”}. Simultaneously your MySQL service is also. Yay! We have successfully created two containers using a docker compose file.
On to the next application. Let’s create a .NET application which interacts with the database and returns a list of strings.
.NET 6.0
We want the .NET application to connect with the database. Secondly, it will fetch data from the database via a GET api and display it in the browser. To do so create a .NET project named dotnet inside our super-app project.
Since writing down all the steps for creating and setting up the project will be quite lengthy for this blog and also to keep this article focused on docker, a separate article is written on how to achieve the above goal. The article provides step by step instructions on creating a .NET 6 application which connects to MySQL database using EF Core. EF Core is an Object-Relation Mapper to communicate between database and code. Note: Ignore the prerequisites section since we already have a MySQL server running in docker.
.NET App Changes
Since the projects will be running in their own containers they can’t communicate via localhost url. Hence in order for the .NET app to communicate with MySQL, the connection string needs to be updated. To do so open the appsettings.json and appsettings.Development.json files and replace localhost with super-app-db for server in DefaultConnection.
Lastly apps within containers don’t need https redirection. Since the https should be handled by the server remove the below line from Program.cs.
app.UseHttpsRedirection(); Docker
Once the project is setup and running it is time to configure it to run inside a docker using Dockerfile and docker compose. In dotnet folder create Dockerfile with below contents.
#Get the SDK image to build and publish the project
FROM mcr.microsoft.com/dotnet/sdk:6.0 AS build-env
WORKDIR /app
# Copy everything
COPY . ./
# Restore as distinct layers
RUN dotnet restore
# Build and publish a release
RUN dotnet publish -c Release -o out
# Build runtime image
FROM mcr.microsoft.com/dotnet/aspnet:6.0
WORKDIR /app
#Copy the build file to the app directory
COPY --from=build-env /app/out .
ENTRYPOINT ["dotnet", "dotnet.dll"]
#Expose the port for communication
EXPOSE 80 Now go back to the docker-compose.yml file and append the below content.
super-app-dotnet:
build: ./dotnet
ports:
- "8080:80" Here we are binding port 8080 of host machine to container’s port 80. That’s all for now. Execute the below command to fire up all the containers.
docker compose up Finally, hit localhost:8080/api/job in a browser. The GET api will fetch the list of jobs from the database.
PYTHON
By now, you might have guessed to create a python folder in our super-app folder 😊. Secondly, create the three files required for our project namely ai-ml.py, requirements.txt and Dockerfile with below contents.
import matplotlib.pyplot as plt
import pandas as pd
from scipy import signal
if __name__ == "__main__":
print("All working good") pandas
scipy
matplotlib # Get the python image
FROM python:3.7.13-slim
# Switch to app directory
WORKDIR /app
# Copy the requirements in to the app
COPY requirements.txt ./
# Install dependencies
RUN pip install --no-cache-dir -r requirements.txt
# Copy everything else
COPY . .
#Run the python script
CMD [ "python", "./ai-ml.py" ] Finally, go back to the docker-compose.yml file and add the below content.
super-app-python:
build: ./python It’s as simple as that. Since it’s just a simple script it will run once and then the container will exit. But the logs for the container will show All working good printed. That’s all for python.
PHP
Setting up PHP with docker is one of the easiest of all. Create two files index.php and Dockerfile as below.
<?php echo "I am running in a container."; ?> # Get the php apache image
FROM php:8.0-apache
# Switch to app directory
WORKDIR /var/www/html
# Copy everything
COPY . .
EXPOSE 80 Lastly, add the below content to docker-compose.yml.
super-app-php:
build: ./php
ports:
- "8000:80" Finally, fire up all the containers again with docker compose up. When you hit http://localhost:8000 a nice message saying “I am running in a container.” will appear.
Conclusion
Docker is a wonderful containerisation tool and it’s been made more powerful with docker compose. It allows to run multiple containers side by side without interfering with each other. It should be in the arsenal of tools in the knowledge base.
I am sure you would enjoy going through some of the articles: