Skip to Content

Supported connectors

Connect Amazon S3

Read files in an S3 bucket as tables.

Before you start

  • You are an Admin or a Developer, or your role has the datasource Connect permission.
  • You have an IAM user with s3:ListBucket on the bucket and s3:GetObject on the dataset folder.

Connect

  1. Go to Data Studio > Datasources and select Connect data source.

    Datasources page with the Connect data source button outlined
    Connect data source is at the top right of the Datasources page.
  2. Select Amazon S3. You can type the name in Search connectors.

    Add datasource page with the Amazon S3 tile outlined
    The Amazon S3 tile on the Add datasource page.
  3. Fill in the form:

    FieldWhat to enter
    Datasource nameHow this datasource appears in lists and pickers, for example Production warehouse.
    S3 RegionThe bucket region. Defaults to us-east-1.
    S3 Access Key ID, S3 Secret Access KeyThe access key of the IAM user.
    S3 Bucket Dataset Folder PathThe folder (prefix) inside the bucket to sync, for example sales/.
    S3 Bucket NameThe bucket name, for example example-sales-bucket.
    Table LevelFile: each file is a table. Folder: each folder is a table.
  4. Select Connect datasource.

Connect Amazon S3 form with the connect button outlined and the Setup guide on the right
Fill in the form, then select the button at the bottom. The Setup guide on the right lists the grants and where each value comes from.

Databrain checks the connection and reads your schema. You see Step 3 of 3 with Model your data. Next, create a datamart.

Amazon S3 is the only connector with a schedule. On the datasource page, select Sync interval to sync every 15 minutes, 30 minutes, 1 hour, 6 hours or 1 day. See Sync a datasource.

Give the IAM user access

The IAM user needs to list the bucket and read the objects under your folder:

s3:GetObject
s3:ListBucket

Where to find the values

  • Region: S3 console > your bucket > Properties.
  • Access key: IAM > the user > Security credentials > Create access key.

If something goes wrong

If the check fails, the datasource stays on the form with Needs credentials. Fix the field the message names and select Connect datasource again.

  • The database did not answer in time. A firewall blocks Databrain. Allow the outbound IP on the database port.
  • The database rejected these credentials. Check the username and password, and that the user can sign in from a remote host.
  • The credentials worked, but the user cannot read the catalogue. Run the grants above.

On this page