For AI agents: a documentation index is available at https://www.mongodb.com/docs/llms.txt — markdown versions of all pages are available by appending .md to any URL path.

Manage Workspace Private Endpoints

A workspace private endpoint connects you to a stream processing workspace over AWS private networking. Atlas Stream Processing generates a private, workspace-specific hostname for each stream processing workspace in the project and maps it to your AWS VPC endpoint (VPCE).

Workspace private endpoints control how clients connect to your stream processing workspace. To instead learn about outbound connections, see Networking.

Important

Workspace private endpoints currently support AWS only. To learn more, see Atlas Stream Processing Limitations.

Interface

To create, view, or delete a workspace private endpoint, you must have the Organization Stream Processing Admin or Project Stream Processing Owner role.

To create a workspace private endpoint, you must have the following:

  • An Atlas project.

  • An AWS user account with an IAM user policy that grants permissions to create, modify, describe, and delete VPC endpoints. To learn more, see the AWS documentation.

  • The AWS CLI installed. Atlas provides AWS CLI commands to find your VPC details and create your VPC endpoint.

  • A VPC in AWS with at least one subnet.

After Atlas Stream Processing provisions a private hostname for your stream processing workspace, you can connect to it with mongosh or VS Code.

The private connection string has this format, where <tenantId> identifies your stream processing workspace and <vpceId> identifies your workspace private endpoint:

mongodb://atlas-stream-<tenantId>-<vpceId>.a-pl.query.mongodb.net/

When you delete a workspace private endpoint, Atlas Stream Processing removes the private hostnames it generated for every stream processing workspace in the project. Clients that connect using those hostnames can no longer reach the affected stream processing workspaces over the private endpoint.