MongoDB Atlas: Free Cluster Setup

What you will learn

By the end of this lesson you will have:

• A free MongoDB database running in the cloud on MongoDB Atlas
• A database user with a username and password
• Your computer’s IP address on the IP access list
• A connection string you can use from mongosh, Compass and your apps

You don’t need a credit card for the free cluster. All you need is a web browser and about 10 minutes.

What is MongoDB Atlas?

MongoDB Atlas is MongoDB’s own cloud service. Instead of installing MongoDB on your computer, Atlas runs it for you on a cloud provider: Amazon Web Services (AWS), Google Cloud or Microsoft Azure. You just connect to it over the internet.

Think of it like this: Installing MongoDB yourself is like cooking at home: you buy the stove, the pans and you clean up. Atlas is like ordering from a good restaurant: the kitchen is someone else’s job, and you just enjoy the meal.

Atlas takes care of the boring but important jobs:

• Keeping the server running and up to date
• Keeping 3 copies of your data on different machines, so one failure doesn’t lose anything
• Turning on security features like passwords and encrypted connections
• Optional backups, monitoring charts, Atlas Search and Vector Search

Local install or Atlas?

You can use both. Many developers learn on a local install and deploy their apps on Atlas.

Note: The Atlas website changes its look from time to time. This lesson describes each step in words. If a button has moved, look for the same words nearby, such as Create, Database Access or Network Access.

Some words you’ll see in Atlas

• Organization: the top level, usually your company or just you.
• Project: a folder inside an organization that holds clusters. You might have one project per app.
• Cluster: your actual MongoDB database server (really a group of servers holding copies of the data).
• Database user: a username and password that apps use to log in to the cluster. This is different from your Atlas website login.
• IP access list: a list of internet addresses that are allowed to connect to your cluster. Everyone else is blocked.

Think of it like this: The cluster is a building. The database user is a door key. The IP access list is the guard at the gate who checks you’re on the guest list. You need both the key and a place on the list to get in.

Step 1: Create an Atlas account

1. Go to mongodb.com/cloud/atlas and choose to try it free or sign up.
2. Sign up with an email address and password, or use a Google or GitHub account.
3. Confirm your email if asked.
4. Atlas may ask a few welcome questions (your goal, your preferred language). Answer them or skip them. They don’t change anything important.

Atlas creates a first organization and project for you automatically.

Step 2: Create a free cluster

1. On your project’s overview page, click Create (sometimes shown as Build a Cluster or Create a cluster).
2. Choose the Free option. In older screens and articles this tier is called M0. It is the same thing.
3. Pick a provider (AWS, Google Cloud or Azure). Any of them is fine for learning.
4. Pick a region close to you, for example Mumbai, Singapore or Frankfurt. Closer means faster. Atlas only lists regions that support free clusters.
5. Give the cluster a name, or keep the default Cluster0. You can’t rename a cluster later, so choose something you like. Letters, numbers and hyphens only.
6. You may see an option to preload a sample dataset. Leaving it on is useful: it gives you real data to play with.
7. Click Create (or Create Deployment).

Note: Some accounts are also asked to choose a “database edition”. For a free cluster the choice makes no difference to how it runs, so the default is fine.

Note: This course uses MongoDB 8.x. Atlas may run your cluster on MongoDB 9.0 or offer 9.0 in a version picker, because new versions reach Atlas first. That’s fine: every lesson in this course works the same way on 9.0.

The free cluster is usually ready within seconds to a few minutes.

What does “free” give you?

The free cluster is meant for learning and small experiments. Its main limits are:

Half a gigabyte sounds small, but it is a lot of text documents. All the shop examples in this course use only a tiny part of it.

Tip: When you outgrow the free cluster, you can scale up to a paid Flex or Dedicated cluster without changing your code.

Step 3: Create a database user

Right after creating the cluster, Atlas usually opens a security quickstart. If it doesn’t, find Database Access in the left sidebar (often grouped under Security) and choose to add a new database user.

1. Choose password authentication.
2. Enter a username, for example asha.
3. Enter a password, or click the button to let Atlas generate a strong one. Copy it somewhere safe now.
4. For the user’s privileges, the default “read and write to any database” is fine for learning.
5. Click Create Database User (or Add User).

Warning: This username and password are what your apps use to log in. Never commit them to GitHub or paste them into public chats. If you leak one, delete the user in Atlas and create a new one.

Tip: To keep things simple, use a password made of letters and numbers only. Characters like @, :, / or % must be specially encoded inside a connection string, which confuses many beginners.

Step 4: Allow your IP address

By default, Atlas blocks every connection. You must add your computer’s internet address to the IP access list.

In the security quickstart, click Add My Current IP Address. Then click Finish and Close.

If you skipped the quickstart:

1. Find Network Access in the left sidebar (often under Security).
2. Click Add IP Address.
3. Click Add Current IP Address, then Confirm.

What about “Allow access from anywhere”?

Atlas also offers an option that adds 0.0.0.0/0, which means “any IP address on the internet”.

• It is handy when your IP changes often (home internet, mobile hotspot) or for some hosting platforms.
• But it means your password is the only thing protecting the database.

For learning, adding your current IP is the safer habit. If your internet provider changes your IP and you suddenly can’t connect, just add the new IP the same way.

Note: Changes to the IP access list can take a minute to apply.

Step 5: Get your connection string

A connection string is like a full postal address for your database. It tells a program where the cluster is and how to log in.

1. Go to your cluster on the Clusters (or Database) page.
2. Click Connect.
3. Choose how you want to connect. Atlas shows options such as Drivers (for apps), Compass, Shell (mongosh) and MongoDB for VS Code.
4. Copy the string Atlas shows you.

It looks something like this:

mongodb+srv://asha:<db_password>@cluster0.abcd123.mongodb.net/?retryWrites=true&w=majority&appName=Cluster0

Let’s read it piece by piece:

You can also add a database name right after the /, before the ?:

mongodb+srv://asha:MyPass123@cluster0.abcd123.mongodb.net/shop?retryWrites=true&w=majority

Now programs using this string start in the shop database.

Step 6: Connect with mongosh

If you have mongosh installed (see Installing MongoDB Community 8), connect like this. Put the address in quotes and let mongosh ask for the password, so it never appears in your terminal history:

mongosh "mongodb+srv://cluster0.abcd123.mongodb.net/" --username asha

mongosh asks for your password. Type it (nothing shows on screen while typing, that’s normal) and press Enter. You’ll see a prompt like:

Atlas atlas-xyz-shard-0 [primary] test>

Try a quick command:

db.runCommand({ ping: 1 })
// → { ok: 1, ... }

If you loaded the sample dataset, list the databases:

show dbs
// → sample_mflix   ...
//   sample_airbnb  ...
//   admin          ...
//   local          ...

Now create the course’s shop database with one product:

use shop
db.products.insertOne({ name: "Notebook", category: "stationery", price: 49, stock: 120, tags: ["paper", "school"] })
// → { acknowledged: true, insertedId: ObjectId("...") }

If you see acknowledged: true, your cloud database is working.

Seeing your data in the Atlas website

On your cluster, open Browse Collections (or Data Explorer). You’ll find the shop database and the products collection with the notebook inside. You can view, edit and delete documents right there.

Step 7 (optional): Use the connection string in an app

Apps read the connection string from an environment variable so the password stays out of the code. In a Node.js project you’d keep it in a .env file:

MONGODB_URI="mongodb+srv://asha:MyPass123@cluster0.abcd123.mongodb.net/shop?retryWrites=true&w=majority"

And your code reads it:

import { MongoClient } from "mongodb";

const client = new MongoClient(process.env.MONGODB_URI);
await client.connect();
console.log("Connected to Atlas!");

Warning: Add .env to your .gitignore file so it never gets pushed to GitHub.

You’ll build a full app in Connecting MongoDB to Your First App.

Bonus: the Atlas CLI

If you prefer the terminal, the Atlas CLI can do all of the steps above with one command, atlas setup. It creates a free cluster, a database user and an IP access entry for you. The website is easier the first time, so we won’t need the CLI in this course.

Troubleshooting connections

Common mistakes

• Using your Atlas website password in the connection string. The connection string needs the database user password you created in Step 3.
• Leaving <db_password> in the string. Replace the whole placeholder, including the angle brackets.
• Forgetting the IP access list. Without it, every connection is silently blocked and simply times out.
• Committing the connection string to GitHub. Keep it in an environment variable or .env file that is ignored by Git.
• Expecting a second free cluster in the same project. Only one free cluster is allowed per project. Create a new project if you need another.

Practice

1. Create a free Atlas cluster and a database user called learner.
2. Add your current IP address and connect with mongosh.
3. In the shop database, insert the Gel Pen product, then check it is there.

// In your terminal (replace the address with your own cluster's):
// mongosh "mongodb+srv://cluster0.abcd123.mongodb.net/" --username learner

use shop

db.products.insertOne({
  name: "Gel Pen",
  category: "stationery",
  price: 15,
  stock: 500,
  tags: ["pen", "school"]
})
// → { acknowledged: true, insertedId: ObjectId("...") }

db.products.find({ name: "Gel Pen" })
// → [ { _id: ObjectId("..."), name: "Gel Pen", category: "stationery", price: 15, ... } ]

Key takeaways

• MongoDB Atlas runs MongoDB for you in the cloud. The Free cluster (formerly M0) costs nothing and never expires.
• To connect you need two things: a database user (key) and your IP on the IP access list (guest list).
• The connection string (mongodb+srv://...) tells programs where the cluster is and who is logging in.
• Keep passwords out of code and Git. Use environment variables.
• The free tier has limits (0.5 GB storage, one per project, pauses after 30 idle days) that are fine for learning.

Next lesson: mongosh: The MongoDB Shell