Skip to content

roadmap.md

Latest commit

 

History

History
92 lines (60 loc) · 3.93 KB

roadmap.md

File metadata and controls

92 lines (60 loc) · 3.93 KB

Torrust Tracker Deployer - Roadmap

This document outlines the development roadmap for the Torrust Tracker Deployer project. Each task is marked with:

  • [ ] - Not completed
  • [x] - Completed

Development Process

When starting work on a new feature:

  1. Create the feature documentation in the docs/features/ folder and commit it
  2. Open an issue on GitHub linking to the feature folder in the repository
  3. Add the new issue as a child issue of the main EPIC issue

Note: See docs/features/README.md for detailed conventions and process guide for creating new features.

Roadmap

1. Add scaffolding for main app

Epic Issue: #2 - Scaffolding for main app

  • 1.1 Setup logging - Issue #3
  • 1.2 Create command torrust-tracker-deployer destroy to destroy an environment
    • It will depend on the environment status. For example, if it has not been provisioned we only have to delete the persistent data.
    • For manual testing we can create envs with the E2E full test (there is an option to keep the infra after testing)
  • 1.3 Refactor extract shared code between testing and production for app bootstrapping
  • 1.4 Improve command to use better abstraction to handle presentation layer
    • types to handle user's views, etc
  • 1.5 Create command torrust-tracker-deployer create to create a new environment
    • We need to decide how the user will provide config values.
    • output using println and eprintln
  • 1.6 Create command torrust-tracker-deployer deploy to run the Provision and Configuration commands (full deployment)
  • 1.7 Add levels of verbosity as described in the UX research

Note: See docs/research/UX/ for detailed UX research that will be useful to implement the features in this section.

2. Add new infrastructure provider: Hetzner

  • 2.1 Add Hetzner provider support

3. Continue adding more application commands

Note: These are internal app layer commands (like ProvisionCommand or ConfigureCommand), not console commands. The approach is to slice by functional services rather than deployment stages - we fully deploy a working stack from the beginning and incrementally add new services.

  • 3.1 Finish ConfigureCommand

    • Add firewall base configuration

  • 3.2 Add ReleaseCommand and RunCommand with slices

    • 3.2.1 Run only a docker compose configuration with hello-world docker image
    • 3.2.2 Add MySQL to docker compose stack
    • 3.2.3 Add Torrust Tracker to docker compose stack
    • 3.2.4 Add Prometheus to docker compose stack
    • 3.2.5 Add Grafana to docker compose stack

    Notes:

    • Each service will require new templates, template Rust wrappers, environment options, etc.

4. Create a docker image for the deployer

  • 4.1 Create docker image for the deployer to use it without needing to install the dependencies (OpenTofu, Ansible, etc)
    • Mount data and build dirs when running it.

5. Add extra console app commands

  • 5.1 torrust-tracker-deployer status - Check environment and service status
  • 5.2 torrust-tracker-deployer test - Run application tests

Note: The test console subcommand is already partially implemented.

6. Add HTTPS support

  • 6.1 Add HTTPS support for HTTP tracker
  • 6.2 Add HTTPS support for tracker API
  • 6.3 Add HTTPS support for Grafana

7. Add backup and disaster recovery

  • 7.1 Implement database backups for MySQL
  • 7.2 Implement configuration backups
  • 7.3 Create recovery procedures documentation

Notes

  • This roadmap will be linked to an EPIC issue on GitHub for tracking progress
  • Each major feature should have corresponding documentation in docs/features/ before implementation begins