Configuring S3 Bucket for Redshift Imports

If you navigate to Settings > Advanced Settings > Integrations, you'll find a setting related to configuring your own S3 bucket to import files into Redshift.

When is this needed?

To import geospatial files seamlessly into your own Amazon Redshift data warehouse, CARTO needs to use an S3 bucket as a middle step, as required by Redshift.

In CARTO Cloud organizations this setting is optional because we provide a pre-configured S3 bucket at all times.

• The import feature will always be enabled for Redshift

• You can use this setting to customize and force the imports to use your own S3 Bucket

In CARTO Self-Hosted organizations you need to configure this setting to be able to import geospatial files into Redshift using CARTO.

Remember that to configure this setting, you previously need to create an S3 bucket in AWS. We recommend you use a fresh new bucket exclusively for this purpose.

Setting up your S3 bucket

First, click on "+ Add" to begin the process. Once inside, tell us the Bucket name and the Region of the S3 Bucket you want to use when importing files.

This S3 bucket needs to be located in the same region as your Amazon Redshift cluster.

Once you've added the name and region, click on Generate policy. This will generate a policy in the code block on the left side, and you should be able to copy it.

Adding the bucket policy in AWS

This step is done in the AWS console. You'll need admin access to the bucket configuration. If you don't have it, just forward an admin these instructions.

The official documentation from AWS covers how to add policies and modify other settings mentioned here.

1. Copy and paste the policy generated by CARTO into the policies of your S3 Bucket in AWS.

2. Make sure ACLs are allowed. This is a must for CARTO to use your S3 bucket.

3. Your S3 Bucket can use encryption but it needs to be managed by AWS.

4. Additionally, you must add a Cross-origin resource sharing (CORS) policy (under the Permissions tab), like the one below. Please replace *.app.carto.com by your specific tenant domain in Self-Hosted deployments.

[ { "AllowedHeaders": [ "*" ], "AllowedMethods": [ "PUT" ], "AllowedOrigins": [ "https://*.app.carto.com" ], "ExposeHeaders": [ "ETag" ], "MaxAgeSeconds": 0 } ]

Validate and save

Once you've finished the configuration in AWS, go back to CARTO. Mark the "I have already added the S3 bucket policy" checkbox as done and click on "Validate and Save"

CARTO will validate your configuration at this point. If we're able to use your S3 bucket to import files into Redshift, we'll take you back to your Settings page.

From this point, CARTO will always use your S3 bucket when importing, and the import feature will be enabled if it wasn't before

You can edit or remove this integration at any time.