Skip to main content
(Applies to dbt v2.0 and later)

Connect Redshift to Fusion Preview

You can configure the Redshift adapter by running dbt init in your CLI or manually providing the profiles.yml file with the fields configured for your authentication type.

The Redshift adapter for Fusion supports the following authentication methods:

  • Password
  • IAM profile

Warehouse permissions

The Redshift database user that dbt Fusion engine uses must be able to run dbt workloads and read catalog metadata used for introspection.

Required Redshift objects

Before connecting, these objects must exist or be accessible:

ObjectPurpose
Cluster (provisioned) or workgroup (serverless)Compute resource
DatabaseTarget database
SchemaTarget schema within the database
UserDatabase user for authentication
IAM role or profile (optional)For IAM-based authentication

Core permissions

The following permissions are required for fundamental dbt features:

PermissionObjectPurpose
USAGESchemaAccess the schema
CREATESchemaCreate tables and views in the schema
SELECTTables or viewsRead data
INSERTTablesInsert data
UPDATETablesUpdate data
DELETETablesDelete data
DROPTables or viewsDrop or replace objects
TRUNCATETablesTruncate tables

Metadata operations

Fusion queries these Redshift system relations:

System relationPurposePermission required
SVV_ALL_COLUMNSColumn metadataSELECT on the system view
pg_classList relationsAccess to system catalog
pg_namespaceSchema informationAccess to system catalog
sys_query_detailSource freshness (last insert time)SELECT on the system view
svv_table_infoList materialized views when the project includes themSELECT on the system view
svv_mv_infoList materialized views when the project includes themSELECT on the system view

Schema management

Conditional permissions for schema management

PermissionObjectWhen required
CREATE SCHEMADatabaseAuto-create schemas

For example SQL grants in Redshift, refer to Redshift permissions.

Configure Fusion

Executing dbt init in your CLI will prompt for the following fields:

  • Host: The hostname of your Redshift cluster
  • User: Username of the account that will be connecting to the database
  • Database: The database name
  • Schema: The schema name
  • Port (default: 5439): Port for your Redshift environment

Alternatively, you can manually create the profiles.yml file and configure the fields. See examples in authentication section for formatting. If there is an existing profiles.yml file, you are given the option to retain the existing fields or overwrite them.

Next, select your authentication method. Follow the on-screen prompts to provide the required information.

Supported authentication types

Use your Redshift user's password to authenticate. You can also manually enter it in plain text into the profiles.yml file configuration.

Example password configuration

profiles.yml
default:
target: dev
outputs:
dev:
type: redshift
port: 5439
database: JAFFLE_SHOP
schema: JAFFLE_TEST
ra3_node: true
method: database
host: ABC123.COM
user: JANE.SMITH@YOURCOMPANY.COM
password: ABC123
threads: 16

More information

Find Redshift-specific configuration information in the Redshift adapter reference guide.

Was this page helpful?

This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.

0
Loading