About access control

This page discusses the two levels of access control for Cloud SQL instances. You must configure both levels of access control before you can manage your instance.

Levels of access control

Configuring access control involves controlling who or what can access the instance. Access control occurs on two levels:

Instance-level access
Instance-level access authorizes access to your Cloud SQL instance from an application or client (running on an App Engine standard environment or externally) or from another Google Cloud service, such as Compute Engine.
Database access
Database access uses server-level roles to control which SQL Server users can access the data in your instance.

Instance-level access

How you configure instance-level access depends on where you are connecting from:
Connection source Access configuration options More information
Google Kubernetes Engine
  • Cloud SQL Auth Proxy Docker image
App Engine standard environment
  • Same project: configure IAM
  • Between projects: configure IAM
flexible environment
  • Same project: preconfigured
  • Between projects: configure IAM
sqlcmd client
  • Cloud SQL Auth Proxy
  • Authorize client IP address
External applications
  • Cloud SQL Auth Proxy
  • Authorize client IP address
Cloud Functions
  • A Cloud SQL instance set up with a public IP.
  • Between projects: also configure IAM
Cloud Run
  • A Cloud SQL instance set up with a public IP.
  • Between projects: also configure IAM
Google Kubernetes Engine
  • Private IP
  • If Public IP, Cloud SQL Auth Proxy is required

Database access

After a user or application connects to a database instance, the user or application must log in with a user or service account. As part of creating a Cloud SQL instance, you set up the default user (root) account. You can also create more users to give you finer-grained control over access to your instance.

For more information, see SQL Server users and Creating and managing SQL Server users.

What's next