Context: I have a SQL Server instance with AWS RDS in a VPC in account A. I have a AWS Data Zone domain created in account B and associated with account A. I created a project in the domain. Now I want to add the RDS instance as data source.
Problem: Currently AWS Data Zone supports two types of Blueprints: AWS Redshift and AWS S3. There seems to be now way at the moment to create a custom Blueprint (e.g. for RDS). But a Blueprint is required to create an environment profile, what is required for the data source.
How do I successfully add the RDS instance as data source (to have the metadata in the catalog as well as be able to manage the permissions via Data Zone)?
As far as I understand it correctly services like AWS RDS are unmanaged assets. I.e. you work with these assets by bringing the metadata to AWS Glue Data Catalog (e.g. using an AWS Glue Crawler) and then import and publish the data from there.
As these kind of resources are unmanaged, it is necessary to implement custom solutions / integrations to read the events from AWS Event Bridge and then execute the corresponding changes in e.g. the AWS RDS instance.
See also https://docs.aws.amazon.com/datazone/latest/userguide/discover-subscribe-consume-data.html