Tracking changes in data in Aurora PostgreSQL

Question

Customer is looking for ways to track changes in their data in Aurora PostgreSQL.

Traditionally they could add a date created, date_modified, username etc columns directly into each table and use it as a reference.

Any suggestions of a way they could cut down on having to do this for each table and instead consume some audit/WAL logs for each row change? Or ideas on how others are tackling this use case?

Accepted Answer

We developed the Database Activity Streams feature for this use case. It's along the lines of pgaudit, but is a managed feature of RDS/Aurora that encrypts events with the customer's key and then streams them to Kinesis. Check it out and let us know what you think!

1. https://docs.aws.amazon.com/AmazonRDS/latest/AuroraUserGuide/DBActivityStreams.html
 2. https://aws.amazon.com/about-aws/whats-new/2019/05/amazon-aurora-with-postgresql-compatibility-supports-database-activity-streams/

Tracking changes in data in Aurora PostgreSQL

Relevant content