Skip to content
Scani docs

Migrate to Tier 2

  1. Back up your database, S3 bucket and existing .env. Keep the same ENCRYPTION_KEY; rotating it makes stored credentials unreadable.
  2. Upgrade the cloud service first, then API and worker to a release supporting processing.v1. Apply the release’s database migrations as usual.
  3. Keep your existing database and S3 configuration. If a previous configuration stored files on a remote operator’s bucket, copy those objects to your own bucket while preserving keys before switching storage routing. Changing an endpoint does not copy files.
  4. Set SCANI_DEPLOYMENT_TIER=2, the cloud URL and your customer cloud key on API and worker. Remove local platform-provider keys and SMTP credentials.
  5. Add docker-compose.tier2.yml to your compose invocation or set COMPOSE_FILE=docker-compose.prod.yml:docker-compose.tier2.yml in .env. Recreate the API and worker. The overlay does not delete old containers; stop your own unused local data-provider and mail catcher explicitly.
  6. Exercise email sign-in, document import, pricing and wallet sync. Check that uploads remain in your bucket and queued work completes.

The installer preserves an existing .env; rerunning it does not select a new tier or replace stored secrets. Edit the routing values yourself when migrating.

To return to Tier 1, set SCANI_DEPLOYMENT_TIER=1, restore the local SCANI_CLOUD_URL and matching local bearer, provide the provider credentials you need, and use the base compose file. Your database and S3 stay where they are.