After getting familiar with Docker, I wanted to play around and try to deploy sample Spring Boot application using PostgresSQL as database server.

What is Docker?

According the official docs Docker is a platform that can be used to build, ship and run applications. In the center of the Docker is the idea of container. Just as the physical ship containers are used to transport goods, the virtual container can be used to ship and deploy any service.

One way of using containers to deploy apps or services is to package it as VM image. Some of the downsides of VM approach are:

  • no guaranty of virtualized environment
  • heavyweight technology (not practical to deploy more than a couple of VMs on developer laptop)
  • slow and IO intensive building and starting process
  • deployment costs on IaaS such as AWS EC2

Very nice introduction to Docker and detailed explanation how it can be used to deploy microservices please read here.

Spring Boot and Docker

Now lets try to build a docker image and run a sample Spring Boot application using containers.

Lets hack

Follows some hands on Docker hacking broken down in six simple steps.

Step 1

Create a Dockerfile to create the Spring boot application Docker image. We will use this image to run containers with the application.

FROM java:8
COPY demo-prod.jar /apps/spring_app/
WORKDIR /apps/spring_app
EXPOSE 8080
CMD ["java", "-jar", "demo-prod.jar"]

The Dockerfile is a simple file that instructs Docker how to build the image. Our file consists of the following instructions:

  • FROM - specifies the starting (base) image from which we build our own image. In our case this is the official java:8 image.
  • COPY - copies the jar of the application in the image file system.
  • WORKDIR - sets the current working directory where the container will run.
  • CMD - instructs Docker what command to run when a container from this image is started.

Step 2

In the second step we build the image running the command:

docker build -t tdelev/spring_app .

giving the image name tdelev/spring_app. Be careful with the . at the end, it is the context directory where Docker builds the image and should only contain our packaged jar application.

Step 3

Next we download already prepared image with Postgres database server. Just run the command and wait…

docker pull postgres

This will download the official postgres image from hub.docker.com.

Step 4

In this step we will create something called data volume container that will be the storage for data from our database.

docker create -v /var/lib/postgresql/data --name spring_app_data postgres:9.4

We use the postgres:9.4 image to create a new volume inside a container of this image at /var/lib/postgresql/data. We choose this location, because this is the default location where postgres stores data from databases. By default this volume is mounted in read-write mode.

Step 5

Now we are ready to start a container with database instance.

docker run --volumes-from spring_app_data \
 --name spring_app_pg \
 -e POSTGRES_USER=demouser -e POSTGRES_PASSWORD=demopass \
 -d -P postgres:9.4 \

This command will start a container named spring_app_pg in detached mode and will set the environment variables POSTGRES_USER and POSTGRES_PASSWORD. It will also mount all volumes from the container spring_app_data that we created in the previous step.

Now we can connect to this database instance using the following command and execute some queries or create databases.

docker run -it --link spring_app_pg:postgres \
--rm postgres sh -c 'exec psql -h "$POSTGRES_PORT_5432_TCP_ADDR" \
-p "$POSTGRES_PORT_5432_TCP_PORT" -U demouser' \

create database demo;

Step 6

In the final step we are ready to run the application. To run, start a container with the following command:

docker run --name spring_app_container \
--link spring_app_pg:spring_app_pg \
-p 8080:8080 \
-d tdelev/spring_app

Here we start a container of the image we created in the Step 2 and use containers linking to link with the database instance container, so our Spring Boot application can connect to the database instance.

We can also connect to the database instance from our host using the port that was automatically exposed from the database instance container. To check the port execute docker ps and check the PORTS column. To connect to the database execute:

psql -h localhost -p 49157 -U demouser --password

where you substitute 49157 with your PORT number.

Presentation

Here you can view my presentation on a JugMK technical session.