Read S3 Parquet using an ECS task identity

How-to guideApplies to: Unified engine deployments

Create the selected AWS credential, bind it to a file view, and read Parquet rows.

Before you start

Use an existing nonempty Parquet dataset and a configured S3 file-source root covering its prefix. A credential does not authorize an arbitrary storage root. The identity needs list/read access to the bucket and actual files, plus KMS decrypt permission when needed. Use the region and HTTPS endpoint for that bucket. Keep the temporary view and its SELECT in the same Workbench workspace. Give the ECS task a reader task role and expose its SDK container-credential endpoint to the credential runtime. An ECS execution role is a different role.

Values to collect

Bucket, Parquet prefix, AWS region, HTTPS S3 endpoint, and the identity-specific values shown below.

Run the SQL

Replace every <placeholder> before running. Use unused names for the resources below. Run the statements in order in Workbench.

CREATE CREDENTIAL recipe_ecs_task_role_reader USING aws
OPTIONS (
  'authentication' = 'container',
  'region' = '<aws-region>'
);

CREATE OR REPLACE TEMPORARY VIEW recipe_ecs_task_role
USING parquet
OPTIONS (
  path 's3://<bucket>/<parquet-prefix>/',
  credential_reference 'recipe_ecs_task_role_reader',
  region '<aws-region>',
  endpoint '<s3-endpoint>',
  allow_http 'false'
);

SELECT * FROM recipe_ecs_task_role LIMIT 5;

Check the result

Expect up to five rows from the Parquet files. Creating or describing the credential alone does not verify S3 access.

Clean up

Remove the temporary view and credentials. The input files stay in place.

DROP VIEW recipe_ecs_task_role;
DROP CREDENTIAL recipe_ecs_task_role_reader;

Last modified on November 4, 2026

Transformation Engine

Learn Laminar

Workloads

SQL reference

HTTP API

Recipes

Operations

Read S3 Parquet using an ECS task identity

Before you start

Values to collect

Run the SQL

Check the result

Clean up

​Before you start

​Values to collect

​Run the SQL

​Check the result

​Clean up

Before you start

Values to collect

Run the SQL

Check the result

Clean up