Example: Using AWS Assume Role Authorization with READ_NOS - Teradata Vantage - Analytics Database

Assume Role gives existing AWS IAM users and service accounts temporary access to AWS resources in other accounts.

The following example shows how you can create an authorization with definer/invoker trusted or simplified credentials and associate it with READ_NOS.

1. Create the DEFINER authorization object with the name S3Definer_auth:

   CREATE AUTHORIZATION S3Definer_auth
   AS DEFINER TRUSTED
   USING
   AUTHSERVICETYPE 'ASSUME_ROLE'
   ROLENAME 'role_name'
   EXTERNALID 'external_id'
   [ DURATION_SECONDS 'duration_in_seconds' ];

   Where:

   role_name is the Amazon Resource Name (ARN) of the role to assume. For example: arn:aws:iam::844102931058:role/STSAssumeRole_TestRole

   external_id is is the external ID which is assuming the role. For example: 3f68bc61-a455-4742-8039-8a03f0600db5

   [OPTIONAL] duration_in_seconds is the duration in seconds of the role session. For example: 7200.

2. Replace the function mapping READ_NOS_FM for the function TD_SYSFNLIB.READ_NOS using the authorization object S3Definer_auth:

   REPLACE FUNCTION MAPPING READ_NOS_FM
   FOR TD_SYSFNLIB.READ_NOS EXTERNAL SECURITY DEFINER TRUSTED S3Definer_auth 
   USING
   BUFFERSIZE ,
   HEADER ,
   ROWFORMAT ,
   MANIFEST ,
   SAMPLE_PERC ,
   RETURNTYPE ,
   LOCATION ,
   ANY IN TABLE;

3. Run READ_NOS_FM function mapping to read data from an S3 CSV bucket at the LOCATION specified. This SQLE call returns the data as a dataset with a single column named payload:

   SELECT payload FROM READ_NOS_FM (
   ON (SELECT cast(NULL as DATASET INLINE LENGTH 64000 STORAGE FORMAT CSV)payload)
   USING
   LOCATION('location')
   ) as d;

   location is the bucket location. For example: /s3/s3.amazonaws.com/stsassumerole-testbucket/BILLING_STATMNT_CHARGE.csv

4. Run READ_NOS_FM function mapping to read all data from an S3 CSV bucket at the LOCATION specified:

   SELECT * FROM READ_NOS_FM (
   USING
   LOCATION('/s3/s3.amazonaws.com/stsassumerole-testbucket/BILLING_STATMNT_CHARGE.csv')
   ) as d;

5. Create the INVOKER authorization object with the name S3Invoker_auth:

   CREATE AUTHORIZATION S3Invoker_auth
   AS INVOKER TRUSTED
   USING
   AUTHSERVICETYPE 'ASSUME_ROLE'
   ROLENAME 'arn:aws:iam::844102931058:role/STSAssumeRole_TestRole'
   EXTERNALID '3f68bc61-a455-4742-8039-8a03f0600db5'
   [ DURATION_SECONDS '7200' ];

6. Replace the function mapping READ_NOS_FM for the function TD_SYSFNLIB.READ_NOS using the authorization object S3Invoker_auth:

   REPLACE FUNCTION MAPPING READ_NOS_FM
   FOR TD_SYSFNLIB.READ_NOS EXTERNAL SECURITY INVOKER TRUSTED S3Invoker_auth 
   USING
   BUFFERSIZE ,
   HEADER ,
   ROWFORMAT ,
   MANIFEST ,
   SAMPLE_PERC ,
   RETURNTYPE ,
   LOCATION ,
   ANY IN TABLE;

7. Run READ_NOS_FM function mapping to read data from an S3 CSV bucket at the LOCATION specified. This SQLE call returns the data as a dataset with a single column named payload:

   SELECT payload FROM READ_NOS_FM (
   ON (SELECT cast(NULL as DATASET INLINE LENGTH 64000 STORAGE FORMAT CSV)payload)
   USING
   LOCATION('/s3/s3.amazonaws.com/stsassumerole-testbucket/BILLING_STATMNT_CHARGE.csv')
   ) as d;

8. Run READ_NOS_FM function mapping to read all data from an S3 CSV bucket at the LOCATION specified:

   SELECT * FROM READ_NOS_FM (
   USING
   LOCATION('/s3/s3.amazonaws.com/stsassumerole-testbucket/BILLING_STATMNT_CHARGE.csv')
   ) as d;

9. Create the authorization object with the name S3_auth:

   CREATE AUTHORIZATION S3_auth
   USING
   AUTHSERVICETYPE 'ASSUME_ROLE'
   ROLENAME 'arn:aws:iam::844102931058:role/STSAssumeRole_TestRole'
   EXTERNALID '3f68bc61-a455-4742-8039-8a03f0600db5'
   [ DURATION_SECONDS '7200' ];

10. Use READ_NOS_FM using the authorization object to read data from an S3 CSV bucket. Note, the READ_NOS_FM command is implicit in the following statement:

    SELECT * FROM (
      LOCATION = 'location'
      AUTHORIZATION = S3_auth) as dt;

    Where location is the bucket location. For example: /s3/s3.amazonaws.com/testuser1-sts-test/csv_default.csv.