I'm trying to deploy my Django app on Heroku. I've created a Heroku Postgres database, and I want to import an existing PostgreSQL database into it from AWS S3. I followed the tutorial in the Heroku docs to the letter. I created a Heroku Postgres database and promoted it to primary db. I followed Heroku's AWS S3 tutorial to set all the config vars so heroku could access my S3 bucket. But when I run heroku pg:backups restore 'https://s3.amazonaws.com/<bucket name>/<database name>' DATABASE_URL, I get an error message saying "An error occurred and your backup didn't finish."
Here is the full output of heroku pg:backups info :
=== Backup info: r012
Database: BACKUP
Started: 2015-07-13 13:43:30 +0000
Finished: 2015-07-13 13:43:30 +0000
Status: Failed
Type: Manual
Backup Size: 0.00B
=== Backup Logs
2015-07-13 13:43:30 +0000: 2015/07/13 13:43:30 aborting: could not write to output stream: Expected HTTP Status 200, received: "403 Forbidden"
2015-07-13 13:43:30 +0000: pg_restore: [custom archiver] could not read from input file: end of file
2015-07-13 13:43:30 +0000: waiting for restore to complete
2015-07-13 13:43:30 +0000: restore done
2015-07-13 13:43:30 +0000: waiting for download to complete
2015-07-13 13:43:30 +0000: download done
Naturally, since this is a 403 error I assumed something was wrong with my authentication. I've run heroku config and triple-checked all my config variables. The AWS_ACCESS_KEY_ID, AWS_SECRET_ACCESS_KEY, and DATABASE_URL environmental variables are all correct, and match what I see on AWS and in the Heroku Postgres GUI.
Why am I seeing this error message?