Render

Render is a unified cloud platform for deploying web services, databases, and background workers with automatic scaling and free SSL.

Why Render?

• Simple pricing: Predictable, straightforward costs
• Built-in databases: Managed PostgreSQL and Redis
• Auto-deploy: Automatic deploys from Git
• Free tier: Limited free services for testing

Architecture

┌─────────────────────────────────────────┐
│              Render Account              │
│                                          │
│  ┌─────────────┐  ┌─────────────────┐   │
│  │ Web Service │  │Background Worker│   │
│  │   (API)     │  │   (Worker)      │   │
│  └──────┬──────┘  └────────┬────────┘   │
│         │                   │            │
│  ┌──────┴───────────────────┴──────┐    │
│  │         Private Network          │    │
│  └──────┬───────────────────┬──────┘    │
│         │                   │            │
│  ┌──────┴──────┐    ┌──────┴──────┐     │
│  │  PostgreSQL  │    │    Redis    │     │
│  └─────────────┘    └─────────────┘     │
└─────────────────────────────────────────┘

Quick Start

1. Create PostgreSQL Database

1. Go to render.com and sign in
2. Click New > PostgreSQL
3. Configure:
   • Name: boards-db
   • Database: boards
   • User: boards
   • Region: Choose closest to users
4. Click Create Database
5. Copy the Internal Database URL

2. Create Redis Instance

1. Click New > Redis
2. Configure:
   • Name: boards-redis
   • Region: Same as database
3. Click Create Redis
4. Copy the Internal Redis URL

3. Deploy API Service

1. Click New > Web Service
2. Select Deploy an existing image from a registry
3. Configure:

Image Settings:

• Image URL: ghcr.io/weirdfingers/boards-backend:latest

Service Settings:

• Name: boards-api
• Region: Same as database
• Instance Type: Starter ($7/month) or higher

Environment Variables:

BOARDS_DATABASE_URL=<internal-database-url>
BOARDS_REDIS_URL=<internal-redis-url>
BOARDS_GENERATOR_API_KEYS={"fal": "your-key", "openai": "your-key"}
BOARDS_AUTH_PROVIDER=none
BOARDS_LOG_FORMAT=json
PORT=8800

Docker Command:

uvicorn boards.api.app:app --host 0.0.0.0 --port 8800

4. Click Create Web Service

4. Deploy Worker Service

1. Click New > Background Worker
2. Select Deploy an existing image from a registry
3. Configure:

Image Settings:

• Image URL: ghcr.io/weirdfingers/boards-backend:latest

Service Settings:

• Name: boards-worker
• Region: Same as database
• Instance Type: Starter or higher

Environment Variables:

BOARDS_DATABASE_URL=<internal-database-url>
BOARDS_REDIS_URL=<internal-redis-url>
BOARDS_GENERATOR_API_KEYS={"fal": "your-key", "openai": "your-key"}
BOARDS_INTERNAL_API_URL=http://boards-api:8800
BOARDS_LOG_FORMAT=json

Docker Command:

boards-worker --log-level info --processes 1 --threads 1

4. Click Create Background Worker

Blueprint (Infrastructure as Code)

Create a render.yaml for one-click deployment:

services:
  # API Service
  - type: web
    name: boards-api
    env: docker
    dockerfilePath: ./Dockerfile
    dockerCommand: uvicorn boards.api.app:app --host 0.0.0.0 --port 8800
    healthCheckPath: /health
    envVars:
      - key: BOARDS_DATABASE_URL
        fromDatabase:
          name: boards-db
          property: connectionString
      - key: BOARDS_REDIS_URL
        fromService:
          name: boards-redis
          type: redis
          property: connectionString
      - key: BOARDS_GENERATOR_API_KEYS
        sync: false
      - key: BOARDS_AUTH_PROVIDER
        value: none
      - key: BOARDS_LOG_FORMAT
        value: json
      - key: PORT
        value: 8800

  # Worker Service
  - type: worker
    name: boards-worker
    env: docker
    dockerfilePath: ./Dockerfile
    dockerCommand: boards-worker --log-level info --processes 1 --threads 1
    envVars:
      - key: BOARDS_DATABASE_URL
        fromDatabase:
          name: boards-db
          property: connectionString
      - key: BOARDS_REDIS_URL
        fromService:
          name: boards-redis
          type: redis
          property: connectionString
      - key: BOARDS_GENERATOR_API_KEYS
        sync: false
      - key: BOARDS_INTERNAL_API_URL
        fromService:
          name: boards-api
          type: web
          property: hostport

databases:
  - name: boards-db
    databaseName: boards
    user: boards

# Note: Redis is a paid feature, add manually if needed

Deploy with blueprint:

1. Push render.yaml to your repository
2. Click New > Blueprint
3. Connect your repository
4. Render creates all services automatically

Environment Groups

Share environment variables across services:

1. Go to Environment Groups
2. Create group boards-shared
3. Add common variables:

BOARDS_GENERATOR_API_KEYS={"fal": "key", "openai": "key"}
BOARDS_AUTH_PROVIDER=jwt
BOARDS_JWT_SECRET=your-secret

4. Link to services in their settings

Custom Domain

1. Go to your API service Settings
2. Under Custom Domains, click Add Custom Domain
3. Enter your domain (e.g., api.boards.example.com)
4. Add the provided CNAME record to your DNS
5. Render automatically provisions SSL

Private Network

Services on Render communicate via private network:

• Use internal URLs (e.g., boards-api:8800)
• No external traffic charges
• Lower latency

Reference internal services:

BOARDS_INTERNAL_API_URL=http://boards-api:8800

Scaling

Horizontal Scaling

1. Go to service Settings
2. Under Scaling, adjust instance count

Vertical Scaling

1. Go to service Settings
2. Change Instance Type:

Type	RAM	CPU	Monthly
Starter	512 MB	0.5 CPU	$7
Standard	2 GB	1 CPU	$25
Pro	4 GB	2 CPU	$85
Pro Plus	8 GB	4 CPU	$175

Deploy Frontend

Option 1: Static Site

1. Click New > Static Site
2. Connect your frontend repository
3. Configure:
   • Build Command: pnpm build
   • Publish Directory: out (for exported Next.js)
4. Add environment variables:

NEXT_PUBLIC_API_URL=https://boards-api.onrender.com
NEXT_PUBLIC_GRAPHQL_URL=https://boards-api.onrender.com/graphql

Option 2: Web Service

For SSR or API routes:

1. Click New > Web Service
2. Connect your frontend repository
3. Configure:
   • Build Command: pnpm build
   • Start Command: pnpm start

Disk Storage

For persistent storage (if using local storage):

1. Go to service Settings
2. Under Disks, click Add Disk
3. Configure:
   • Name: storage
   • Mount Path: /app/data/storage
   • Size: 10 GB (minimum)

Monitoring

Logs

View logs in dashboard:

1. Click on service
2. Go to Logs tab

Or use the CLI:

render logs --service boards-api

Metrics

Render provides basic metrics:

• CPU usage
• Memory usage
• Request count
• Response times

Health Checks

Configure health checks:

1. Go to service Settings
2. Set Health Check Path: /health
3. Render restarts unhealthy instances automatically

CI/CD

Automatic Deploys

Enable in service settings:

1. Go to Settings
2. Under Build & Deploy, enable Auto-Deploy
3. Deploys trigger on push to connected branch

Deploy Hooks

Use webhooks for custom CI:

1. Go to service Settings
2. Find Deploy Hook URL
3. POST to this URL to trigger deploy

Cost Estimation

Service	Type	Monthly
API	Starter	$7
Worker	Starter	$7
PostgreSQL	Starter	$7
Redis	Starter	$10
Total		~$31

Scale up as needed. Free tier available for static sites.

Troubleshooting

Build Failures

1. Check build logs in dashboard
2. Verify Dockerfile/commands are correct
3. Test build locally first

Service Not Starting

1. Check logs for errors
2. Verify environment variables
3. Check Docker command syntax

Database Connection

1. Verify using internal URL
2. Check database is in same region
3. Test connection:

render shell boards-api
python -c "from boards.db import engine; print(engine.connect())"

Slow Cold Starts

Render spins down free/starter instances. To prevent:

• Upgrade to Standard or higher
• Use health check pings to keep warm

Next Steps

• Storage Configuration - Configure S3 for production storage
• Authentication - Set up auth providers
• Monitoring - Add external monitoring