Create a geodatabase on Amazon RDS for PostgreSQL or Aurora PostgreSQL

Use the Create Database Connection geoprocessing tool to connect to an existing instance of Amazon Relational Database Service (RDS) for PostgreSQL or Amazon Aurora (PostgreSQL-compatible edition). Use the Enable Enterprise Geodatabase geoprocessing tool to create a geodatabase in the database. Run both tools from an ArcGIS client installed on an Amazon Elastic Compute Cloud (EC2) instance running in the same region as the database instance.

You need to create your own enterprise geodatabase in the following situations:

• You use ArcGIS Enterprise deployment tools for Amazon Web Services (AWS) to deploy an ArcGIS Server site, but you want to customize the name or location of the geodatabase you register with your ArcGIS Server site.
• You use ArcGIS Enterprise deployment tools for Amazon Web Services (AWS) to deploy an ArcGIS Server site, but you have multiple departments that maintain their own discrete data and require their own geodatabases.
• You configure a stand-alone ArcGIS Server site on AWS using the Amazon Web Services Management Console. In this case, you must create your own databases, sde and data owner login roles, sde and data owner schemas, and geodatabases, and register the geodatabases with the ArcGIS Server site.

The following steps summarize what you need to create and the format these objects must take to be used with ArcGIS on AWS.

Follow instructions in PostgreSQL and AWS documentation to create all objects other than the geodatabase (steps 1 through 9).

1. Create the database instance and database, and load the PostGIS extension to the database.
2. Create an sde login role in PostgreSQL and grant it the rds_superuser role.

   Use all lowercase characters for the username; mixed and uppercase character names are not supported with ArcGIS.

3. Create a schema named sde in the database, and make the sde login the owner of the sde schema.

   The schema name must match the username exactly to be used with ArcGIS.

4. Create a login role that will own data. Grant this login privileges to create tables and views in the database.

   Use all lowercase characters for the login role; mixed and uppercase character names are not supported with ArcGIS.

5. Create a schema for this new login role.

   The schema name must exactly match the login role name to be used with ArcGIS. Make the new login role the owner of the schema.

6. Grant this new login usage on the sde schema.
7. Grant the sde login usage on the new login's schema.
8. Set the search path on the database so it searches the connected user's schema first ("$user"), searches the public schema second, and searches the sde schema last.
9. Grant all privileges on the database to the sde login.
10. Run the Create Database Connection geoprocessing tool from ArcGIS Pro or ArcGIS Server and connect as the sde user.

    You will use the resultant database connection file (.sde) as input in the next step.

11. Run the Enable Enterprise Geodatabase geoprocessing tool from the ArcGIS client.

    You cannot use the Create Enterprise Geodatabase geoprocessing tool; you must use the Enable Enterprise Geodatabase tool.

12. Run the Create Database Connection geoprocessing tool from the ArcGIS client to connect to the database as the login you created in step 4 to load data.

    Tip:

    Rename the resultant connection file to differentiate it from the one you created to connect as the sde user.

    You will use the resultant database connection file (.sde) as input in the next step.

13. Use this new connection file to register your geodatabase with the ArcGIS Server site running on AWS.

    For federated ArcGIS Server sites, add a database data store item in the portal.

    For stand-alone ArcGIS Server sites, register from ArcGIS Server Manager.