Metadata Access Control - AWS Lake Formation

Metadata Access Control

For access control for Data Catalog resources, the following discussion assumes fine-grained access control with Lake Formation permissions and coarse-grained access control with IAM policies.

There are two distinct methods for granting Lake Formation permissions on Data Catalog resources:

  • Named resource access control – With this method, you grant permissions on specific databases or tables by specifying database or table names. The grants have this form:

    Grant permissions to principals on resources [with grant option].

    With the grant option, you can allow the grantee to grant the permissions to other principals.

  • Tag-based access control – With this method, you assign one or more LF-tags to Data Catalog databases, tables, and columns, and grant permissions on one or more LF-tags to principals. Each LF-tag is a key-value pair, such as department=sales. A principal that has LF-tags that match the LF-tags on a Data Catalog resource can access that resource. This method is recommended for data lakes with a large number of databases and tables. It's explained in detail in Overview of Lake Formation Tag-Based Access Control.

The permissions that a principal has on a resource is the union of the permissions granted by both the methods.

The following table summarizes the available Lake Formation permissions on Data Catalog resources. The column headings indicate the resource on which the permission is granted.

Catalog Database Table
CREATE_DATABASE CREATE_TABLE ALTER
ALTER DROP
DROP DESCRIBE
DESCRIBE SELECT*
INSERT*
DELETE*

For example, the CREATE_TABLE permission is granted on a database. This means that the principal is allowed to create tables in that database.

The permissions with an asterisk (*) are granted on Data Catalog resources, but they apply to the underlying data. For example, the DROP permission on a metadata table enables you to drop the table from the Data Catalog. However, the DELETE permission granted on the same table enables you to delete the table's underlying data in Amazon S3, using, for example, a SQL DELETE statement. With these permissions, you also can view the table on the Lake Formation console and retrieve information about the table with the AWS Glue API. Thus, SELECT, INSERT, and DELETE are both Data Catalog permissions and data access permissions.

When granting SELECT on a table, you can add a filter that includes or excludes one or more columns. This permits fine-grained access control on metadata table columns, limiting the columns that users of integrated services can see when running queries. This capability is not available using just IAM policies.

There is also a special permission named Super. The Super permission enables a principal to perform every supported Lake Formation operation on the database or table on which it is granted. This permission can coexist with the other Lake Formation permissions. For example, you can grant Super, SELECT, and INSERT on a metadata table. The principal can perform all supported actions on the table, and when you revoke Super, the SELECT and INSERT permissions remain.

For details on each permission, see Lake Formation Permissions Reference.

Important

To be able to see a Data Catalog table created by another user, you must be granted at least one Lake Formation permission on the table. If you are granted at least one permission on the table, you can also see the table's containing database.

You can grant or revoke Data Catalog permissions using the Lake Formation console, the API, or the AWS Command Line Interface (AWS CLI). The following is an example of an AWS CLI command that grants the user datalake_user1 permission to create tables in the retail database.

aws lakeformation grant-permissions --principal DataLakePrincipalIdentifier=arn:aws:iam::111122223333:user/datalake_user1 
 --permissions "CREATE_TABLE" --resource '{ "Database": {"Name":"retail"}}'

The following is an example of a coarse-grained access control IAM policy that complements fine-grained access control with Lake Formation permissions. It permits all operations on any metadata database or table.

{
    "Version": "2012-10-17",
    "Statement": [
        {
            "Effect": "Allow",
            "Action": [
                "glue:*Database*",
                "glue:*Table*",
                "glue:*Partition*"
            ],
            "Resource": "*"
        }
    ]
}

The next example is also coarse-grained but somewhat more restrictive. It permits read-only operations on all metadata databases and tables in the Data Catalog in the designated account and Region.

{  
    "Version": "2012-10-17",
    "Statement": [
        {
            "Effect": "Allow",
            "Action": [
                "glue:GetTables",
                "glue:SearchTables",
                "glue:GetTable",
                "glue:GetDatabase", 
                "glue:GetDatabases"
            ],
            "Resource": "arn:aws:glue:us-east-1:111122223333:*"
        } 
    ]   
}

Compare these policies to the following policy, which implements IAM-based fine-grained access control. It grants permissions only on a subset of tables in the customer relationship management (CRM) metadata database in the designated account and Region.

{  
    "Version": "2012-10-17",
    "Statement": [
        {
            "Effect": "Allow",
            "Action": [
                "glue:GetTables",
                "glue:SearchTables",
                "glue:GetTable",
                "glue:GetDatabase", 
                "glue:GetDatabases"
            ],
            "Resource": [
                "arn:aws:glue:us-east-1:111122223333:catalog",
                "arn:aws:glue:us-east-1:111122223333:database/CRM",
                "arn:aws:glue:us-east-1:111122223333:table/CRM/P*"
            ]
        } 
    ]   
}

For more examples of coarse-grained access control policies, see Lake Formation Personas and IAM Permissions Reference.