> ## Documentation Index
> Fetch the complete documentation index at: https://docs.thanx.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Redshift Serverless

> Configuring your Redshift Serverless destination.

## Prerequisites

* [ ] If your Redshift security posture requires IP whitelisting, have the data-syncing service's static IP available during the following steps. It will be required in Step 2.
* [ ] By default, Redshift Serverless authentication uses role-based access. You will need the trust policy prepopulated with the data-syncing service's identifier to grant access. It should look similar to the following JSON object with a proper service account identifier:

```json theme={null}
{
  "Version": "2012-10-17",
  "Statement": [
    {
      "Effect": "Allow",
      "Action": [
        "sts:AssumeRoleWithWebIdentity"
      ],
      "Principal": {
        "Federated": "accounts.google.com"
      },
      "Condition": {
        "StringEquals": {
          "accounts.google.com:oaud": "<some_organization_identifier>",
          "accounts.google.com:sub": "<some_service_account_identifier>"
        }
      }
    }
  ]
}
```

<Note>
  **Network allowlisting**

  Cloud-hosted (US): `35.192.85.117/32`

  Cloud-hosted (EU): `104.199.49.149/32`

  If private-cloud or self-hosted, contact your Thanx representative for the static egress IP.
</Note>

<Note>
  **How authentication works**

  Two distinct identities are involved: an AWS IAM role in your account (created in Step 3), which holds the S3 staging permissions and is allowed to call `redshift-serverless:GetCredentials` on your workgroup, and a Redshift database user (created in Step 1) derived from that IAM identity.

  The database username format depends on the authentication method: `IAMR:<role-name>` for an IAM role (the final segment of the role ARN), or `IAM:<user-name>` for an AWS user with an HMAC access key.
</Note>

## Step 1: Create the database user in Redshift Serverless

1. Connect to Redshift Serverless using an SQL client.
2. Create the user, quoting the identifier because of the colon (replace `<role-name>` with the final segment of your IAM role ARN, or use the `IAM:<user-name>` format for HMAC access key authentication).

```sql theme={null}
CREATE USER "IAMR:<role-name>" PASSWORD DISABLE;
```

<Note>
  **If the user already exists**

  Redshift Serverless auto-creates this user on the first connection attempt. If a prior connection was already made, the `CREATE USER` statement fails with "user already exists" — skip it and proceed to the grant below.
</Note>

3. Grant user `create` and `temporary` privileges on the database. `create` allows the service to create new schemas and `temporary` allows the service to create temporary tables.

```sql theme={null}
GRANT CREATE, TEMPORARY ON DATABASE <database> TO "IAMR:<role-name>";
```

<Note>
  **The schema will be created during the first sync**

  The schema name supplied as part of Step 4 will be created during the first connection. It does not need to be created manually in the destination ahead of time.
</Note>

<Warning>
  **If the `schema` already exists**

  By default, the service creates a new schema based on the destination configuration. If you prefer to create the schema yourself before connecting the destination, you must ensure that the writer user has the proper permissions on the schema, using `GRANT ALL ON schema <schema> TO "IAMR:<role-name>";`

  Once you've provided the `GRANT ALL` permission on the schema, you can safely remove the `CREATE` permission on the database (but you must retain the `TEMPORARY` permission on the database).
</Warning>

## Step 2: Whitelist connection

1. In the Redshift console, click **Workgroups**, and make a note of the **workgroup** name.
2. Select the workgroup you would like to connect.
3. In the **General information** pane, make note of the **Endpoint** details. You may need to use the **copy** icon to copy the full details to discover the full endpoint and port number.

![](https://storage.googleapis.com/prequel_docs/images/redshift-endpoint-details.png "redshift endpoint details.png")

4. Click the **Properties** tab.
5. Scroll down to the **Network and security settings** section.
6. In the VPC security group field, select a security group to open it.

![](https://storage.googleapis.com/prequel_docs/images/redshift-vpc-security-groups.png "redshift vpc s groups.png")
7\. In the Security Groups window, click **Inbound rules**.
8\. Click **Edit inbound rules**.
9\. In the Edit the Inbound rules window, follow the steps below to create custom TCP rules for the static IP:

* Select **Custom TCP** in the drop-down menu.
* Enter your Redshift port number. (likely `5439`)
* Enter the **static IP**.
* Click **Add rule**.

## Step 3: Create a staging bucket

### Create staging bucket

1. Navigate to the S3 service page.
2. Click Create bucket.
3. Enter a **Bucket name** and modify any of the default settings as desired. Note: **Object Ownership** can be set to "**ACLs disabled**" and **Block Public Access settings for this bucket** can be set to "**Block all public access**" as recommended by AWS. Make note of the Bucket name and AWS Region.
4. Click **Create bucket**.

<Note>
  **Optional: add a short retention lifecycle policy**

  You may configure a lifecycle rule on the staging bucket to automatically delete objects older than 2 days, since the bucket is not used to persist data. In the bucket's **Management** tab, click **Create lifecycle rule**, and set an expiration action for current versions of objects with a 2-day age. Transfer logic automatically cleans up files after each transfer completes, so this step is optional.
</Note>

### Create policy

1. Navigate to the **IAM** service page, click on the **Policies** navigation tab, and click **Create policy**.
2. Click the JSON tab, and paste the following policy, being sure to replace `BUCKET_NAME` with the name of the bucket chosen above, and `REGION_NAME`, `ACCOUNT_ID`, and `WORKGROUP_NAME_OR_ID` with the proper Redshift Serverless values.
   * **Note**: the first bucket permission in the list applies to `BUCKET_NAME` whereas the second permission applies only to the bucket's contents — `BUCKET_NAME/*` — an important distinction.

```json JSON policy theme={null}
{
    "Version": "2012-10-17",
    "Statement": [
        {
            "Effect": "Allow",
            "Action": [
                "redshift-serverless:GetCredentials"
            ],
            "Resource": [
                "arn:aws:redshift-serverless:REGION_NAME:ACCOUNT_ID:workgroup/WORKGROUP_NAME_OR_ID"
            ]
        },
        {
            "Effect": "Allow",
            "Action": "s3:ListBucket",
            "Resource": "arn:aws:s3:::BUCKET_NAME"
        },
        {
            "Effect": "Allow",
            "Action": [
                "s3:PutObject",
                "s3:GetObject",
                "s3:DeleteObject"
            ],
            "Resource": "arn:aws:s3:::BUCKET_NAME/*"
        }
    ]
}
```

<Note>
  **KMS encryption (optional)**

  If your S3 staging bucket uses KMS encryption (CMK), add the following statement to the `Statement` array of your IAM policy to allow data encryption and decryption with your KMS key. Encryption with SSE-C is not currently supported.

  ```json theme={null}
  {
    "Effect": "Allow",
    "Action": [
      "kms:GenerateDataKey",
      "kms:Decrypt"
    ],
    "Resource": "arn:aws:kms:REGION_NAME:ACCOUNT_ID:key/KEY_ID"
  }
  ```

  Replace `REGION_NAME`, `ACCOUNT_ID`, and `KEY_ID` with your values.
</Note>

3. Click through to the **Review** step, choose a **name** for the policy, for example, `transfer-service-policy` (this will be referenced in the next step), add a description, and click **Create policy**.

<Tabs>
  <Tab title="IAM role (recommended)">
    ### Create role

    1. Navigate to the **IAM** service page.
    2. Navigate to the **Roles** navigation tab, and click **Create role**.
    3. Select **Custom trust policy** and paste the provided trust policy (from the prerequisite) to allow AssumeRole access to this role. Click **Next**.
    4. Add the permissions policy created above, and click **Next**.
    5. Enter a **Role name**, for example, `transfer-role`, and click **Create role**.
    6. Once successfully created, search for the created role in the Roles list, click the role name, and make a note of the **ARN** value. The final segment of this ARN must match the role name used for the database user in Step 1.
  </Tab>

  <Tab title="AWS user with HMAC access key">
    Role based authentication is the preferred authentication mode for Redshift Serverless based on AWS recommendations. However, HMAC Access Key ID & Secret Access Key is an alternative authentication method that can be used if preferred.

    1. Navigate to the **IAM** service page.
    2. Navigate to the **Users** navigation tab, and click **Add users**.
    3. Enter a **User name** for the service, for example, `transfer-service`, click **Next**. Under **Select AWS access type**, select the **Access key - Programmatic access** option. Click **Next: Permissions**.
    4. Click the **Attach existing policies directly** option, and search for the name of the policy created in the previous step. Select the policy, and click **Next: Tags**.
    5. Click **Next: Review** and click **Create user**.
    6. In the **Success** screen, record the **Access key ID** and the **Secret access key**.
  </Tab>
</Tabs>

## Step 4: Add your destination

Securely share your **workgroup**, **host**, **database**, your chosen **schema**, **IAM role ARN**, and **staging bucket details** with us to complete the connection.

## Permissions checklist

* Redshift Serverless database user `"IAMR:<role-name>"` (or `"IAM:<user-name>"` for access key authentication) exists and has `CREATE` and `TEMPORARY` on the database. If you pre-created the schema, ensure `GRANT ALL ON SCHEMA <schema> TO "IAMR:<role-name>"`.
* IAM role trust policy allows the data-syncing service to assume the role.
* IAM policy includes:
  * `redshift-serverless:GetCredentials` on your target workgroup ARN.
  * S3 `ListBucket` on `arn:aws:s3:::BUCKET_NAME`.
  * S3 `GetObject`, `PutObject`, `DeleteObject` on `arn:aws:s3:::BUCKET_NAME/*`.
* Network allowlisting (if enforced) permits the egress IP/CIDR for the Redshift port (typically 5439).

## FAQ

<AccordionGroup>
  <Accordion title="How is the Redshift Serverless connection secured?">
    We use role-based authentication with your AWS IAM Role. We assume your role to obtain short-lived database credentials, and network access can be constrained by allowlisting the static egress IPs noted above.
  </Accordion>

  <Accordion title="Why is an S3 bucket required?">
    Redshift's high-throughput path loads data from S3 using `COPY`. We stage files briefly in your bucket to maximize throughput and reliability. Files are cleaned up after load.
  </Accordion>

  <Accordion title="Why is redshift-serverless:GetCredentials needed on the workgroup?">
    In Serverless, temporary database credentials are issued per workgroup, so granting this action allows the assumed role to obtain ephemeral credentials without long-lived secrets.
  </Accordion>

  <Accordion title="What are the `oaud` vs `sub` IDs used for?">
    These are identity claims used in the IAM trust policy when federating from GCP to AWS. `sub` uniquely identifies our Google principal in federation. `oaud` is an additional claim used to bind role assumption to your organization.
  </Accordion>

  <Accordion title="Why am I getting authentication errors with Redshift Serverless?">
    Common causes:

    * Missing or incorrect permission on `redshift-serverless:GetCredentials` (ensure it targets the correct workgroup ARN and region/account).
    * Trust policy mismatch (our principal is not permitted to assume your role).
    * Using a provisioned cluster ARN and `redshift:GetClusterCredentials` instead of the Serverless workgroup permission.
    * Database privileges granted to the wrong user.
    * Propagation delay: IAM changes can take a few minutes to apply. Retry after 5-10 minutes.
  </Accordion>

  <Accordion title="Do I need to pre-create the schema?">
    No. The schema provided in the destination configuration is created automatically on first sync. If you pre-create it, grant `ALL` on the schema to the writer user, and you may remove the database-level `CREATE` permission (retain `TEMPORARY`).
  </Accordion>
</AccordionGroup>