Skip to content

Latest commit

 

History

History
45 lines (33 loc) · 1.81 KB

deployment.md

File metadata and controls

45 lines (33 loc) · 1.81 KB
Raw

Initial setup

Pre-read epic stack deployment

First add secrets (via flyctl or Fly web)

  • INTERNAL_COMMAND_TOKEN
  • RESEND_API_KEY
  • SESSION_SECRET

To create a new app, use the following two commands:

fly launch --name evs-production --org opportunity-hack -r sjc
fly consul attach --app evs-production
fly deploy --app evs-production

Deployment

First time setup

Most of the necessary configuration for this application to be continuously deployed is already encoded in the Dockerfile, the fly.toml configuration file, and .github/workflows/deploy.yml.

This setup should be sufficient for pushing application changes and database migrations to deployments without much manual intervention; however, for the application to function properly, the database must be seeded with some role and permission data that is not defined by the database schema, as well as an initial admin user.

To do this:

  1. Run fly ssh console -a <YOUR_APP_NAME> to connect to your deployment

  2. After connecting to your server, run the command sqlite3 /litefs/data/sqlite.db < /evs/prisma/deployment_seed.sql

This sql script will create instructor, admin, and necessary volunteer user roles, as well as an initial admin user with the username "admin" and a hardcoded password au1pui2OBjJdTzGS5c0F. It will also create a default signup secret with the value of "horses are cool" At this point, you should create a user of your own, use this default admin account to give your new user account admin rights through the UI, and then delete the default admin account.

If you want to customize the seeding process, see the instructions here: https://github.com/epicweb-dev/epic-stack/blob/main/docs/deployment.md#seeding-production

At this point you should be all set!