impress/deploy
Emi Matchu 9ba94f9f4b chore: document legacy openneo_id migrations and update references
This commit completes the database consolidation cleanup by documenting
the historical migrations and updating all references to reflect the
single-database architecture.

Changes:
- db/openneo_id_migrate/README.md: Created comprehensive documentation
  explaining the history of the separate database and why these migrations
  are preserved but no longer runnable
- db/openneo_id_schema.rb: Deleted (no longer needed)
- README.md: Updated to reflect single-database architecture
  - Removed mentions of "two databases"
  - Updated "OpenNeo ID Database" section to "Authentication Architecture"
  - Added reference to historical context in db/openneo_id_migrate/README.md
- deploy/setup.yml: Removed openneo_id database creation and privileges
  for future deployments
- db/migrate/20240401124200_increase_username_length.rb: Updated comment
  to note this was historically paired with an openneo_id migration

The codebase now fully reflects the consolidated single-database architecture.
The legacy migration files in db/openneo_id_migrate/ are preserved for
historical reference only.

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>
2025-11-02 07:02:43 +00:00
..
files Set up deployment inside devcontainer 2025-11-02 04:02:06 +00:00
ansible.cfg Speed up deploys with Ansible's pipelining option 2024-09-06 12:22:28 -07:00
deploy.yml Oops, re-enable setting as default upon deploy 2025-10-31 03:34:02 +00:00
inventory.cfg Upgrade to Ruby 3.4.5, Rails 8.0.2 2025-07-23 20:39:27 -07:00
README Create setup.yml deploy script 2023-10-23 19:05:09 -07:00
rollback.yml Upgrade to Ruby 3.4.5, Rails 8.0.2 2025-07-23 20:39:27 -07:00
setup.yml chore: document legacy openneo_id migrations and update references 2025-11-02 07:02:43 +00:00

Dress to Impress is deployed to a VPS server. We use this Ansible Playbook to
automate the environment setup!

We expect to be deploying to Ubuntu 20.04 LTS, initially with nothing
installed. The user you deploy with should have sudoers access. That should be
all it takes!

First, run `yarn deploy:setup` in the app root, to run the `setup.yml`
playbook. This will prompt you for your root password, to set up system
dependencies. It should be safe to re-run this, including if you add a new
dependency to the playbook, because the steps are non-destructive and Ansible
will skip steps that are already satisfied.

Then, to deploy a new version of the app, run `yarn deploy`. This will build
the app from the code on your machine, then send the source and build output
to the remote machine, and switch it to be the new production version. Nice!

Note that the setup script references a file named `production.env`, which is
gitignored because it contains sensitive information, like database passwords.
You should create a `production.env` file in the local `deploy/files`
directory, to be copied to the remote server and used as its environment
variables.