How it works
Overview
This document provides an overview of how Atlas Schema Monitoring works. It describes the high level architecture, key capabilities and features.
Architecture
Atlas Cloud never has direct access to your database, instead it uses a middleman, the Atlas agent, to connect to your
database instead. In order for this to work, the agent needs to be installed somewhere with network connectivity to the
database, usually within the same VPC as the database. In addition, the agent should have outbound connectivity to your cloud account (e.g.,https://<tenant>.atlasgo.cloud).
The agent then starts polling Atlas Cloud for work. Once assigned a task, it connects to the database and executes the task, e.g. "take a snapshot" and then reports back the result to Atlas Cloud. The Agent does not read or report back any user data, only meta information about the database schema.
Key Capabilities
Documentation
Schema Monitoring provides a set of tools and features to help you manage and monitor your database schema effectively. In addition to a nice ERD visualization of your database schema, you get auto generated Go-style like documentation of your database schema.
Schema Changelog
Once configured, the Atlas agent periodically checks the database schema and if there is a change, reports it back to the cloud where you can see detailed information about the changes detected.
Drift Detection
If you have also set up deployment reporting to Atlas Cloud, Atlas can check if the actual state of your schema matches the expected one if there is a drift, e.g. because someone manually made a change to the database schema. If Atlas detects such a drift, you will see it in the Change Log.
