Skip to main content
Skip to main content

S3

In this document, you’ll learn how to install the S3 plugin on your Medusa backend and use it for storage.

Overview​

To upload and manage file assets in Medusa, you need a file service plugin responsible for hosting the files. Without a file service plugin, you will face issues while working with Medusa, such as when uploading images for products.

Medusa provides three different options to handle your file storage. This document focuses on using S3 to store images and files uploaded to the Medusa backend.

Prerequisites​

Medusa Backend​

A Medusa backend is required to be set up before following along with this document. You can follow the quickstart guide to get started in minutes.

Required Accounts​

You need to create an AWS account to follow along with this documentation.

Create S3 Bucket​

On your AWS Console, search for S3 in the search box at the top. Then, choose the first result you see which should be S3 under the Services category.

Enter S3 in the search box and choose S3 from the result in the Services category

Then, on the new page that opens, click on Create Bucket button at the top right of the Buckets table.

Click on the Create bucket button

The Create Bucket form will open. In the General Configuration section enter a name for the bucket and choose a region for the bucket. Both of the values of these fields are important as you’ll use them throughout the documentation.

Enter bucket name and choose region in the General Configuration section

Next, in the Object Ownership section, choose ACLs enabled. Then, two radio buttons will show below it. Choose Bucket owner preferred.

In the Object Ownership section choose ACLs enabled, then choose Bucket owner preferred for the Object ownership field

Then, in the “Block Public Access settings for this bucket” section, uncheck the “Block all public access” checkbox. This shows a warning message at the bottom of the section with another checkbox. Check the checkbox to ensure you understand that objects in the bucket are publicly accessible.

Uncheck Block all public access and all fields under it, and check the checkbox field "I acknowledge that the current settings might result in this bucket and the objects within becoming public"

You can leave the rest of the fields in the form as is and scroll down to the end of the page. Then, click on the Create Bucket button.

Manage Bucket Policies​

On the page of the bucket you just created, click on the Permissions tab. Then, scroll down until you find the Bucket policy section. Click on Edit in that section.

Find the bucket policy section which should be empty and click on the Edit button

In the Edit Bucket Policy page, enter the following in the field:

{
  "Version": "2012-10-17",
  "Id": "Policy1397632521960",
  "Statement": [
    {
      "Sid": "Stmt1397633323327",
      "Effect": "Allow",
      "Principal": {
        "AWS": "*"
      },
      "Action": "s3:GetObject",
      "Resource": "arn:aws:s3:::<YOUR_BUCKET_NAME>/*"
    }
  ]
}

Make sure to replace <YOUR_BUCKET_NAME> with the name of the bucket you created.

Once you’re done, scroll down and click on the Save changes button.

User Permissions​

Your user must have the AmazonS3FullAccess policy attached to it. You can refer to this guide to learn how to add a policy if necessary.

Obtain Access Keys​

You must obtain access keys for your user as you’ll use them to integrate the S3 plugin in Medusa with your bucket. To obtain the Access Key ID and the Secret Access Key, check out this guide.

Install the S3 Plugin​

In the directory of your Medusa backend, run the following command to install the S3 Plugin:

npm install medusa-file-s3

Then, in medusa-config.js, add to the plugins array the following new item:

medusa-config.js
const plugins = [
  // ...
  {
    resolve: `medusa-file-s3`,
    options: {
        s3_url: process.env.S3_URL,
        bucket: process.env.S3_BUCKET,
        region: process.env.S3_REGION,
        access_key_id: process.env.S3_ACCESS_KEY_ID,
        secret_access_key: process.env.S3_SECRET_ACCESS_KEY,
        cache_control: process.env.S3_CACHE_CONTROL,
        // optional
        download_file_duration:
          process.env.S3_DOWNLOAD_FILE_DURATION,
        prefix: process.env.S3_PREFIX,
    },
  },
]

Where:

  1. s3_url (required) is a string indicating the URL to your bucket. It’s in the form https://<BUCKET_NAME>.s3.<REGION>.amazonaws.com, where <BUCKET_NAME> is the name of the bucket and the <REGION> is the region the bucket is created in. If you’re not sure which region, on your bucket’s page on S3 click on Properties. You can then find the region under AWS Region. Make sure to only copy the code (for example, us-east-1).
  2. bucket (required) is a string indicating the name of the bucket you created.
  3. region (required) is a string indicating the region code of your bucket. For example, us-east-1.
  4. access_key_id (required) is a string indicating the Access Key ID that you created for your user.
  5. secret_access_key (required) is a string indicating the Secret Access Key that you created for your user.
  6. cache_control (default: max-age=31536000) is a string indicating the value for caching the objects in a web browser or CDN network. For example: cache object for 10 hours, max-age=36000
  7. download_file_duration (optional) is a string indicating the expiry duration for a download URL. For example, you can set the value to 3600 if you want the download URL to expire in 1 hour.
  8. prefix (optional) is a string indicating a prefix to apply on stored file names.

Make sure to also set these values in your environment variables as well:

S3_URL=<YOUR_BUCKET_URL>
S3_BUCKET=<YOUR_BUCKET_NAME>
S3_REGION=<YOUR_BUCKET_REGION>
S3_ACCESS_KEY_ID=<YOUR_ACCESS_KEY_ID>
S3_SECRET_ACCESS_KEY=<YOUR_SECRET_ACCESS_KEY>
S3_CACHE_CONTROL=<YOUR_CACHE_CONTROL>
S3_DOWNLOAD_FILE_DURATION=<YOUR_DOWNLOAD_FILE_DURATION_AMOUNT>
S3_PREFIX=<YOUR_BUCKET_PREFIX>
Warning

If you have multiple storage plugins configured, the last plugin declared in the medusa-config.js file will be used.

Add AWS Configurations​

You can pass additional AWS configurations, such as customUserAgent, in the plugin's options under the property aws_config_object. This property is an object that accepts AWS Configurations as its properties.

For example:

medusa-config.js
const plugins = [
  // ...
  {
    resolve: `medusa-file-s3`,
    options: {
        s3_url: process.env.S3_URL,
        bucket: process.env.S3_BUCKET,
        aws_config_object: {
          customUserAgent: process.env.S3_CUSTOM_AGENT,
        },
    },
  },
]

Make sure to define S3_CUSTOM_AGENT in your environment variables first.

Test the S3 Plugin​

Run your Medusa backend with the following command:

npx medusa develop

Then, you can either test the plugin using the REST APIs or using the Medusa Admin.

On the Medusa Admin, create a new product and, in the Images section, upload an image then click Save. If the integration was successful, the product image will be uploaded successfully.

An image is successfully uploaded on the Medusa Admin

You can also check that the image was uploaded on the S3 bucket’s page.

Image is now available in the S3 Bucket

Next.js Starter Template Configuration​

If you’re using a Next.js Starter Template, you need to add an additional configuration that adds the S3 bucket domain name into the configured images’ domain names. This is because all URLs of product images will be from the S3 bucket.

If this configuration is not added, you’ll receive the error "next/image Un-configured Host”.

In next.config.js add the following option in the exported object:

next.config.js
const { withStoreConfig } = require("./store-config")

// ...

module.exports = withStoreConfig({
  // ...
  images: {
    remotePatterns: [
      // ...
      {
        protocol: "https",
        hostname:"<BUCKET_NAME>.s3.<REGION>.amazonaws.com",
      },
      // ...
    ],
  },
})

Where:

  • <BUCKET_NAME> is the name of the S3 bucket you’re using
  • <REGION> is the region of the S3 bucket (for example, eu-west-1). If your S3 URL doesn't use region in it, you may omit it to be instead <BUCKET_NAME>.s3.amazonaws.com.

Troubleshooting​

Error: AccessControlListNotSupported: The bucket does not allow ACLs

See Also​

Was this section helpful?
© 2024 Medusa, Inc. All rights reserved.