This post outlines a straightforward process for migrating your Heroku Postgres database to Hatchbox Postgres. The following steps assume both your Heroku and Hatchbox applications are already working with their respective databases.

Prerequisites

  • Both applications have their databases up and running.
  • You have SSH access to your Hatchbox server.
  • Your Hatchbox app is configured with the necessary environment variables.

Step-by-Step Migration

Step 1: Retrieve the Heroku Database URL

Get your Heroku database URL and save it as an environment variable in Hatchbox:

HEROKU_URL=$(heroku config:get DATABASE_URL --app yourapp)

Step 2: SSH into Your Hatchbox Server

Log in to your Hatchbox server via SSH:

ssh deploy@your-hatchbox-domain

Step 3: Change Directory to Your Hatchbox App

Navigate to your Hatchbox application’s current release directory:

cd ~/YOUR-APP/current

Step 4: Load the Environment Variables

Load the environment variables required by your Hatchbox app:

eval "$(/home/deploy/.asdf/bin/asdf vars)"

Step 5: Run pgsync to Migrate the Data

Execute the pgsync command to start the migration:

bundle exec pgsync --to "$DATABASE_URL" --from "$HEROKU_URL" --debug --to-safe

If you encounter issues related to foreign keys or constraints, consider deferring those constraints:

bundle exec pgsync --to "$DATABASE_URL" --from "$HEROKU_URL" --defer-constraints --debug --to-safe

Final Notes

  • Data Integrity: Always review any error messages carefully. Issues related to foreign keys or other constraints might require adjusting your data or modifying constraint settings.
  • Testing: Once migration completes, perform thorough testing to confirm that your application operates correctly on Hatchbox Postgres.

This concise set of steps should help you migrate your database effectively.