Grey Haven Cloudflare Deployment

Deploy TanStack Start applications to Cloudflare Workers using GitHub Actions, Doppler for secrets, and Wrangler CLI.

Deployment Architecture

TanStack Start on Cloudflare Workers

• SSR: Server-side rendering with TanStack Start server functions
• Edge Runtime: Global deployment on Cloudflare’s edge network
• Database: PostgreSQL (PlanetScale) with connection pooling
• Cache: Cloudflare KV for sessions, R2 for file uploads
• Secrets: Managed via Doppler, injected in GitHub Actions

Core Infrastructure

• Workers: Edge compute (TanStack Start)
• KV Storage: Session management
• R2 Storage: File uploads and assets
• D1 Database: Edge data (optional)
• Queues: Background jobs (optional)

Wrangler Configuration

Basic wrangler.toml

name = "grey-haven-app"
main = "dist/server/index.js"
compatibility_date = "2025-01-15"
node_compat = true

[vars]
ENVIRONMENT = "production"
DATABASE_POOL_MIN = "2"
DATABASE_POOL_MAX = "10"

# KV namespace for session storage
[[kv_namespaces]]
binding = "SESSIONS"
id = "your-kv-namespace-id"
preview_id = "your-preview-kv-namespace-id"

# R2 bucket for file uploads
[[r2_buckets]]
binding = "UPLOADS"
bucket_name = "grey-haven-uploads"
preview_bucket_name = "grey-haven-uploads-preview"

# Routes
routes = [
  { pattern = "app.greyhaven.studio", zone_name = "greyhaven.studio" }
]

Environment-Specific Configs

• Development: `wrangler.toml

…(truncated)