Deployment & Infrastructure

The EGAC platform runs entirely on Cloudflare’s free tier. There is no server to maintain — the Worker, D1 database, KV namespace, and cron triggers are all Cloudflare-managed.

Infrastructure overview

First-time setup

1. Create Cloudflare resources

# Create D1 database
wrangler d1 create egac

# Create KV namespace
wrangler kv:namespace create "KV"

Copy the generated IDs into wrangler.toml:

[[d1_databases]]
binding = "DB"
database_name = "egac"
database_id = "YOUR-D1-ID-HERE"

[[kv_namespaces]]
binding = "KV"
id = "YOUR-KV-ID-HERE"

2. Run database migrations

# Apply all migrations to production
wrangler d1 execute egac --remote --file=db/migrations/0001_initial_schema.sql
wrangler d1 execute egac --remote --file=db/migrations/0002_membership_forms.sql
wrangler d1 execute egac --remote --file=db/migrations/0003_update_academy_waitlist_template.sql
wrangler d1 execute egac --remote --file=db/migrations/0004_coach_app.sql

3. Set secrets

wrangler secret put RESEND_API_KEY   # From resend.com dashboard
wrangler secret put ADMIN_TOKEN      # Strong random string for admin access
wrangler secret put CRON_SECRET      # Strong random string for cron auth

4. Deploy

npm run build
wrangler pages deploy dist --project-name=egac

Or connect to GitHub for automatic deploys on every push.

wrangler.toml reference

name = "egac"
compatibility_date = "2024-09-23"
pages_build_output_dir = "dist"

[vars]
APP_ENV = "production"
EMAIL_FROM = "no-reply@updates.eastgrinsteadac.co.uk"

[[d1_databases]]
binding = "DB"
database_name = "egac"
database_id = "6a053384-20ee-4fee-bf73-aa29c1779fe3"

[[kv_namespaces]]
binding = "KV"
id = "28ecc4df31b44e5e92cc4730c1870829"

Cron trigger configuration

Cron triggers must be configured in the Cloudflare Pages dashboard (not in wrangler.toml for Pages projects):

Workers & Pages → egac → Settings → Functions → Cron Triggers

Local development

npm install
npm run dev       # Astro dev server on http://localhost:4321

The dev server uses a local D1 database. Run migrations against local D1:

wrangler d1 execute egac --local --file=db/migrations/0001_initial_schema.sql
# ... repeat for other migrations

Seed with test data:

wrangler d1 execute egac --local --file=db/seed.sql

In development, use Authorization: Bearer dev for admin and cron routes.

CI/CD with GitHub

1. Connect the repository to Cloudflare Pages via the dashboard
2. Set build command: npm run build
3. Set output directory: dist
4. Add the CLOUDFLARE_API_TOKEN environment variable to GitHub Secrets for Wrangler CLI use in CI

Every push to main triggers a production build. Pull requests get automatic preview deployments at unique URLs.

Database migrations

Migrations are plain SQL files in db/migrations/, numbered sequentially. There is no migration runner — apply them manually via Wrangler in order:

# Check what's been applied (no built-in tracker — use naming convention)
wrangler d1 execute egac --remote --command "SELECT name FROM sqlite_master WHERE type='table'"

# Apply a new migration
wrangler d1 execute egac --remote --file=db/migrations/0005_new_feature.sql

Environment variables summary