Running a Postgres server on Docker

Through this article, you will get a Postgres server in a Docker container up and running, expose that server to the outside World and use it from a Jupyter Notebook.

Before getting started, you need to download Docker for your OS, create an account and log in.

Docker Desktop for Mac and Windows | Docker

For this experiment, we will be using Visual Studio Code. You will need to open a new terminal window and work from there.

1. To begin, we will pull your preferred Postgres image, and in this example we will use the latest available.

Docker pull postgres:latest

We can confirm our image has been downloaded by typing

Docker Images

2. Now it’s time to run our Postgres image. I will name it randydb-postgres, and for the password I will use “chosenpass.” Lastly, I want to expose my server to PORT 5432 in order to access it from outside the container.

Docker run — name randydb-postgres -e POSTGRES_PASSWORD=chosenpass -d -p 5432:5432 postgres

You can see our running images by typing the command

Docker ps

3. To go inside our container from our terminal use the following command:

Docker exec -it randydb-postgres bash

4. To directly access Postgres use this command:

psql -U postgres

You can get a list of all commands by typing \h

You can quit postgres by typing \q

5. We can finally create our Database and start querying it.

CREATE DATABASE chosen_dbname

To confirm the creation of our DB you can type \l and see a list of all databases.

6. Switch into your new database with the following command

\c chosen_dbname

You are now connected to database “chosen_dbname” as user “postgres”

From there you can change user credentials, make changes to your database, etc.

The last step is to connect from the outside such as a Jupyter Notebook.

I will be using the psycogpg2 library which you can install by running:

!pip install psycogpg2 in your Notebook

# Create a connection to your server

conn = psycopg2.connect(“host=127.0.0.1 dbname=chosen_dbname user=postgres password=chosenpass”)

# Get your cursor to the connection and query away!

cur = conn.cursor()

Congratulations, you just created your first dockerized server!