When not needing the SQL function and updater Lambda stack (for Postgres connection), how best to remove?

[chrisl777]

Amplify CLI Version

14.2.3

Question

Question, we successfully connected to a Postgres database (external to AWS) from our Gen 2 schema, following the docs here: https://docs.amplify.aws/react-native/build-a-backend/data/connect-to-existing-data-sources/connect-postgres-mysql-database/

I see that Amplify automatically creates and manages the SQL function and updater Lambda for connecting to the Postgres database, in a CloudFormation stack.

We no longer need the connection to Postgres and want to remove it from our schema. What is the best way to get rid of the CloudFormation stack for the SQL function and updater Lambda? I don't see anything in the docs about what do to tear down the stack.

[pahud]

Hi @chrisl777,

Thanks for reaching out! This is a good question, and I can see the documentation doesn't currently cover the removal workflow for SQL data sources.

To remove the Postgres connection and clean up the associated CloudFormation stack:

1. Remove the SQL data source configuration from your schema in amplify/data/resource.ts. This includes:

   • Any models that were connected to the Postgres database
   • The SQL data source strategy configuration
   • Any related connection configuration

2. Deploy your changes:

   # For sandbox/development
   npx ampx sandbox
   

3. Verify cleanup in the AWS Console:

   • Go to CloudFormation and confirm the SQL-related stack was deleted
   • Check Lambda to ensure the SQL function and updater Lambda were removed
   • Verify IAM roles associated with the SQL connection are cleaned up

Please let me know if it works for you.

[chrisl777]

@pahud Thank you! Per your steps above, it appears that the SQL stack is properly getting removed. So I think we're good to close this issue.

Btw, the Postgres connection worked very well. Perhaps in the future, there could be "first class" Postgres support for Amplify data schemas (i.e. specify models and have it provision the backend infrastructure for the GraphQL API).