Digital Ocean Deployment Guide

Complete guide for deploying PutPlace to Digital Ocean droplets with automated provisioning, deployment, and management.

Overview

This deployment method provides:

Simple setup: Single command deployment
Cost-effective: Starting at $6/month
Full control: Root access to your server
No vendor lock-in: Standard Ubuntu + Python stack
MongoDB included: Runs on the same droplet or use Atlas
S3 storage support: Optional AWS S3 for file storage with automatic credentials deployment
SSL/HTTPS ready: Certbot integration for Let’s Encrypt

Cost Comparison

Platform	Monthly Cost	Notes
Digital Ocean (1GB)	$6	Good for testing/development
Digital Ocean (2GB)	$12	Recommended for production
Digital Ocean (4GB)	$24	High traffic production
AWS AppRunner	$25+	+ NAT Gateway ($32) if using Atlas
Heroku	$25+	Hobby tier

Quick Start

Prerequisites

Digital Ocean account: Sign up at https://www.digitalocean.com/
Digital Ocean CLI (doctl): Install and authenticate
SSH key: Add to your Digital Ocean account
API token: Create in Digital Ocean dashboard

Installation

# Install doctl (macOS)
brew install doctl

# Authenticate doctl
doctl auth init

# Add your SSH key to Digital Ocean (if not already done)
doctl compute ssh-key list
# If empty, create one:
ssh-keygen -t ed25519 -C "your_email@example.com"
doctl compute ssh-key import my-key --public-key-file ~/.ssh/id_ed25519.pub

# Set Digital Ocean API token
export DIGITALOCEAN_TOKEN="your_token_here"

Simplified Deployment (Recommended)

The easiest way to deploy with environment-specific shortcuts:

# Step 1: Configure production environment (one-time setup)
invoke configure-prod --mongodb-url="mongodb+srv://user:pass@cluster.mongodb.net/"

# This creates:
# - IAM user: putplace-prod
# - S3 bucket: putplace-prod
# - Config file: ppserver-prod.toml
# - AWS credentials in: aws_credentials_output/

# Step 2: Deploy to Digital Ocean
invoke deploy-do-prod --create

# This will:
# 1. Create a 1GB droplet (putplace-prod) in Frankfurt
# 2. Wait for it to boot (~60 seconds)
# 3. Install Python, nginx, uv (skips MongoDB - using Atlas)
# 4. Install PutPlace from PyPI
# 5. Copy AWS credentials to droplet
# 6. Set up systemd service with AWS profile
# 7. Configure nginx reverse proxy
# 8. Start the application
#
# Total time: ~5 minutes (faster without MongoDB!)

Available environment shortcuts:

configure-dev / deploy-do-dev - Development
configure-test / deploy-do-test - Testing
configure-prod / deploy-do-prod - Production

Manual Deployment (Full Control)

For advanced use cases with full parameter control:

# Configure manually
invoke configure --envtype=prod --setup-iam \
    --storage-backend=s3 --s3-bucket=putplace \
    --mongodb-url="mongodb+srv://..."

# Deploy manually
invoke deploy-do --create --droplet-name=putplace-prod \
    --storage-backend=s3 --s3-bucket=putplace-prod \
    --aws-profile=putplace-prod --mongodb-url="..."

Output:

=== PutPlace Digital Ocean Deployment ===

Creating droplet: putplace-prod
  Region: fra1
  Size: s-1vcpu-1gb
  Image: ubuntu-22-04-x64
  Using SSH keys: 12345678

✓ Droplet created successfully!
  ID: 334455667
  IP: 165.22.123.45

Waiting for SSH to be available on 165.22.123.45...
✓ SSH is ready

Provisioning server at 165.22.123.45...
  Running provisioning script (this may take 5-10 minutes)...
✓ Server provisioned successfully

Deploying application from https://github.com/jdrumgoole/putplace.git...
✓ Application deployed successfully

Setting up systemd service...
✓ Systemd service configured and started

Setting up nginx reverse proxy...
✓ Nginx configured successfully

=== Deployment Complete! ===

Application URL: http://165.22.123.45
API Docs: http://165.22.123.45/docs
SSH Access: ssh root@165.22.123.45

Next steps:
1. Configure admin user and API keys
2. Set up SSL certificate (if using domain)
3. Configure firewall rules
4. Set up monitoring and backups

Deployment Options

Deploy to Existing Droplet

If you already have a droplet:

# Deploy by droplet name
invoke deploy-do --droplet-name=putplace-prod

# Deploy by IP address
invoke deploy-do --ip=165.22.123.45

Deploy with Custom Domain

# Deploy with domain name
invoke deploy-do --create --droplet-name=putplace-prod --domain=api.example.com

# After deployment, point your domain's A record to the droplet IP
# Then enable SSL:
ssh root@165.22.123.45 'certbot --nginx -d api.example.com'

Custom Droplet Size and Region

# Deploy with 2GB RAM (recommended for production)
invoke deploy-do --create --droplet-name=putplace-prod --size=s-1vcpu-2gb

# Deploy in a different region
invoke deploy-do --create --droplet-name=putplace-prod --region=nyc1

# Common regions:
#   fra1  - Frankfurt, Germany
#   lon1  - London, UK
#   nyc1  - New York, USA
#   sfo3  - San Francisco, USA
#   sgp1  - Singapore
#   tor1  - Toronto, Canada

# Common sizes:
#   s-1vcpu-1gb   - $6/month  (512 MB transfer)
#   s-1vcpu-2gb   - $12/month (2 TB transfer) ← Recommended
#   s-2vcpu-4gb   - $24/month (4 TB transfer)
#   s-4vcpu-8gb   - $48/month (5 TB transfer)

Deploy with AWS S3 Storage

Recommended: Use environment shortcuts

# Configure production with S3 (one-time setup)
invoke configure-prod --mongodb-url="mongodb://localhost:27017"

# Deploy to production
invoke deploy-do-prod --create

Manual approach with full control:

# First, generate AWS credentials (one-time setup)
invoke configure --envtype=prod --setup-iam \
    --storage-backend=s3 \
    --s3-bucket=putplace

# Deploy with S3 storage and AWS credentials
invoke deploy-do --create \
    --droplet-name=putplace-prod \
    --storage-backend=s3 \
    --s3-bucket=putplace-prod \
    --aws-profile=putplace-prod

# Deploy to existing droplet with S3
invoke deploy-do \
    --droplet-name=putplace-prod \
    --storage-backend=s3 \
    --s3-bucket=putplace-prod \
    --aws-profile=putplace-prod

See the “Using AWS S3 for File Storage” section below for detailed configuration.

Deploy with MongoDB Atlas and S3 Storage

Recommended: Use environment shortcuts

# Configure production with Atlas and S3 (one-time setup)
invoke configure-prod --mongodb-url="mongodb+srv://user:pass@cluster.mongodb.net/"

# Deploy to production (skips MongoDB installation)
invoke deploy-do-prod --create

Manual approach with full control:

# Complete cloud deployment (no local storage)
invoke deploy-do --create \
    --droplet-name=putplace-prod \
    --mongodb-url="mongodb+srv://user:pass@cluster.mongodb.net/" \
    --storage-backend=s3 \
    --s3-bucket=putplace-prod \
    --aws-profile=putplace-prod

Quick Updates

After initial deployment, use the update command for fast code updates:

# Update by droplet name
invoke update-do --droplet-name=putplace-prod

# Update by IP
invoke update-do --ip=165.22.123.45

# Update specific branch
invoke update-do --ip=165.22.123.45 --branch=develop

This pulls the latest code, updates dependencies, and restarts the service. Takes ~30 seconds vs ~10 minutes for full deployment.

Management Commands

SSH Access

# SSH by droplet name
invoke ssh-do --droplet-name=putplace-prod

# SSH by IP
invoke ssh-do --ip=165.22.123.45

View Logs

# View access logs (last 50 lines)
invoke logs-do --droplet-name=putplace-prod

# View error logs
invoke logs-do --ip=165.22.123.45 --error

# Follow logs in real-time
invoke logs-do --ip=165.22.123.45 --error --follow

# Manual log viewing via SSH
ssh root@165.22.123.45 'tail -f /var/log/putplace/error.log'
ssh root@165.22.123.45 'tail -f /var/log/putplace/access.log'

Service Management

# SSH into droplet
invoke ssh-do --droplet-name=putplace-prod

# Check service status
systemctl status putplace

# Restart service
systemctl restart putplace

# View systemd logs
journalctl -u putplace -f

# Check nginx status
systemctl status nginx

# Test nginx configuration
nginx -t

Post-Deployment Configuration

1. Create Admin User

The application creates an admin user automatically on first startup. Check the credentials:

ssh root@165.22.123.45 'cat /tmp/putplace_initial_creds.txt'

Or set custom credentials before first startup:

ssh root@165.22.123.45

# Edit .env file
cd /opt/putplace/putplace
nano .env

# Add:
PUTPLACE_ADMIN_USERNAME=admin
PUTPLACE_ADMIN_PASSWORD=your-secure-password
PUTPLACE_ADMIN_EMAIL=admin@example.com

# Restart service
systemctl restart putplace

2. Enable SSL/HTTPS (Recommended)

If you have a domain pointed to your droplet:

ssh root@165.22.123.45

# Run certbot (interactive)
certbot --nginx -d api.example.com

# Follow prompts to:
# 1. Enter email address
# 2. Agree to terms
# 3. Choose redirect HTTP to HTTPS (recommended)

# Auto-renewal is configured automatically
# Test renewal:
certbot renew --dry-run

Your API will now be available at https://api.example.com.

3. Configure Firewall (UFW)

ssh root@165.22.123.45

# Enable firewall
ufw allow OpenSSH
ufw allow 'Nginx Full'
ufw enable

# Check status
ufw status

4. Configure Environment Variables

Edit /opt/putplace/putplace/.env:

ssh root@165.22.123.45
cd /opt/putplace/putplace
nano .env

Important settings:

# MongoDB (local)
MONGODB_URL=mongodb://localhost:27017
MONGODB_DATABASE=putplace

# Or use MongoDB Atlas
MONGODB_URL=mongodb+srv://user:pass@cluster.mongodb.net/putplace

# Registration
ALLOW_REGISTRATION=false  # Disable public registration

# Logging
LOG_LEVEL=INFO  # DEBUG, INFO, WARNING, ERROR

After changes:

systemctl restart putplace

Architecture

Server Layout

/opt/putplace/
  └── ppserver.toml          # Application configuration

/var/putplace/
  └── storage/               # Local file storage (if not using S3)

/root/.aws/                  # AWS credentials (if using S3)
  ├── credentials            # AWS access keys
  └── config                 # AWS region/profile config

/var/log/putplace/
  ├── access.log             # Application access logs
  └── error.log              # Application error logs

/etc/systemd/system/
  └── putplace.service       # Systemd service file

/etc/nginx/
  └── sites-available/
      └── putplace           # Nginx configuration

Services

MongoDB (mongod.service)
- Port: 27017 (localhost only)
- Data: /var/lib/mongodb
- Logs: /var/log/mongodb
PutPlace (putplace.service)
- Port: 8000 (localhost only)
- Workers: 4 (uvicorn)
- Logs: /var/log/putplace/
Nginx (nginx.service)
- Ports: 80 (HTTP), 443 (HTTPS)
- Reverse proxy to port 8000
- SSL termination

Network Flow

Internet
  ↓
Nginx (port 80/443)
  ↓
PutPlace (ppserver, port 8000)
  ↓
  ├─→ MongoDB (port 27017, localhost or Atlas)
  └─→ File Storage (Local disk or AWS S3)

Using MongoDB Atlas Instead

If you prefer MongoDB Atlas over local MongoDB:

1. Create Atlas Cluster

Sign up at https://www.mongodb.com/cloud/atlas
Create a free M0 cluster or paid cluster
Add droplet IP to IP whitelist
Create database user
Get connection string

2. Update Configuration

ssh root@165.22.123.45
cd /opt/putplace/putplace

# Edit .env
nano .env

# Update MongoDB URL
MONGODB_URL=mongodb+srv://username:password@cluster.mongodb.net/putplace

# Restart service
systemctl restart putplace

# Optional: Disable local MongoDB to save resources
systemctl stop mongod
systemctl disable mongod

This saves ~200MB RAM by not running MongoDB locally.

Using AWS S3 for File Storage

PutPlace supports storing file content in AWS S3 instead of local disk storage. This section shows how to configure S3 storage with automatic AWS credentials deployment.

Prerequisites

AWS Account with S3 access
IAM Users Created: Use pp_configure with --setup-iam to create IAM users and S3 buckets
Credentials Generated: Located in ./aws_credentials_output/ directory

1. Generate AWS Credentials (One-Time Setup)

On your local machine, create AWS IAM users, S3 buckets, and credentials for your environment.

Recommended: Use environment shortcuts

# Production environment (creates putplace-prod IAM user and bucket)
invoke configure-prod --mongodb-url="mongodb+srv://user:pass@cluster.mongodb.net/"

# Development environment (creates putplace-dev IAM user and bucket)
invoke configure-dev --mongodb-url="mongodb://localhost:27017"

# Test environment (creates putplace-test IAM user and bucket)
invoke configure-test --mongodb-url="mongodb://localhost:27017"

Manual approach with full control:

# Generate IAM users, S3 buckets, and credentials
invoke configure --envtype=prod --setup-iam \
    --storage-backend=s3 \
    --s3-bucket=putplace \
    --aws-region=eu-west-1 \
    --mongodb-url="mongodb+srv://user:pass@cluster.mongodb.net/"

Both approaches create:

IAM user: putplace-{envtype} (e.g., putplace-prod)
S3 bucket: putplace-{envtype} (e.g., putplace-prod)
Credentials in: ./aws_credentials_output/
- aws_credentials (AWS credentials format)
- aws_config (AWS config format)
- .env.{envtype} (environment variable format)

What gets created:

aws_credentials_output/
├── aws_credentials          # Multi-profile credentials file
│   [putplace-dev]
│   aws_access_key_id = AKIAIOSFODNN7EXAMPLE
│   aws_secret_access_key = wJalrXUtnFEMI/K7MDENG/bPxRfiCYEXAMPLEKEY
│   [putplace-prod]
│   ...
├── aws_config              # Multi-profile config file
│   [profile putplace-dev]
│   region = eu-west-1
│   ...
├── .env.dev                # Environment-specific credentials
├── .env.prod
└── .env.test

2. Deploy with S3 Storage and AWS Credentials

The deployment script automatically copies AWS credentials to the droplet and configures the application to use them.

Recommended: Use environment shortcuts

# Deploy to new production droplet (reads ppserver-prod.toml)
invoke deploy-do-prod --create

# Deploy to existing production droplet
invoke deploy-do-prod

# Deploy to dev/test environments
invoke deploy-do-dev --create
invoke deploy-do-test --create

Manual approach with full control:

# Deploy to new droplet with S3 storage
invoke deploy-do --create \
    --droplet-name=putplace-prod \
    --storage-backend=s3 \
    --s3-bucket=putplace-prod \
    --aws-profile=putplace-prod

# Deploy to existing droplet with S3 storage
invoke deploy-do \
    --droplet-name=putplace-prod \
    --storage-backend=s3 \
    --s3-bucket=putplace-prod \
    --aws-profile=putplace-prod \
    --aws-credentials-dir=./aws_credentials_output

What happens during deployment:

Generates ppserver.toml locally with S3 configuration
Copies AWS credentials from ./aws_credentials_output/ to droplet:
- aws_credentials → /root/.aws/credentials
- aws_config → /root/.aws/config
- Sets proper permissions (600) for security
Configures systemd service with AWS_PROFILE=putplace-prod environment variable
Application automatically uses S3 for file storage with the specified profile

3. Verify S3 Configuration

After deployment, verify S3 is working:

# Check AWS credentials are configured
ssh root@165.22.123.45 'cat /root/.aws/credentials'
ssh root@165.22.123.45 'aws s3 ls s3://putplace-prod/'

# Check service environment
ssh root@165.22.123.45 'systemctl show putplace | grep AWS_PROFILE'

# Upload a test file via API
curl -X POST http://165.22.123.45/put_file \
  -H "Content-Type: application/json" \
  -d '{
    "filepath": "/test/file.txt",
    "hostname": "test-host",
    "ip_address": "192.168.1.1",
    "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855"
  }'

# Verify file is in S3
ssh root@165.22.123.45 'aws s3 ls s3://putplace-prod/e3/b0/'

4. Alternative: Manual AWS Credentials Setup

If you prefer to configure AWS credentials manually without using --setup-iam:

# SSH into droplet
ssh root@165.22.123.45

# Create AWS credentials directory
mkdir -p /root/.aws

# Create credentials file
cat > /root/.aws/credentials << 'EOF'
[putplace-prod]
aws_access_key_id = YOUR_ACCESS_KEY_ID
aws_secret_access_key = YOUR_SECRET_ACCESS_KEY
EOF

# Create config file
cat > /root/.aws/config << 'EOF'
[profile putplace-prod]
region = eu-west-1
output = json
EOF

# Set proper permissions
chmod 600 /root/.aws/credentials /root/.aws/config

# Update systemd service to use profile
nano /etc/systemd/system/putplace.service

# Add under [Service]:
Environment="AWS_PROFILE=putplace-prod"

# Reload and restart
systemctl daemon-reload
systemctl restart putplace

5. Environment-Specific Deployments

Deploy different environments with appropriate credentials.

Recommended: Use environment shortcuts

The shortcut tasks automatically configure the correct IAM user, S3 bucket, and droplet name for each environment:

# Development environment
invoke configure-dev --mongodb-url="mongodb://localhost:27017"
invoke deploy-do-dev --create

# Test environment
invoke configure-test --mongodb-url="mongodb://localhost:27017"
invoke deploy-do-test --create

# Production environment
invoke configure-prod --mongodb-url="mongodb+srv://user:pass@cluster.mongodb.net/"
invoke deploy-do-prod --create

Manual approach with full control:

# Development environment
invoke deploy-do --create \
    --droplet-name=putplace-dev \
    --storage-backend=s3 \
    --s3-bucket=putplace-dev \
    --aws-profile=putplace-dev

# Production environment
invoke deploy-do --create \
    --droplet-name=putplace-prod \
    --storage-backend=s3 \
    --s3-bucket=putplace-prod \
    --aws-profile=putplace-prod

S3 Storage Benefits

Advantages:

Unlimited storage: No disk space limits
Durability: 99.999999999% (11 nines) durability
Redundancy: Automatic replication across availability zones
Cost-effective: Pay only for what you use (~$0.023/GB/month)
Scalability: Handles any number of files
Security: Encryption at rest and in transit

When to use S3:

Storing large amounts of file content
Multi-region deployments (shared storage)
High durability requirements
Compliance/audit requirements

When to use local storage:

Small deployments with limited file storage needs
Lower latency requirements (files on same server)
Cost optimization for very small datasets (<10GB)
Development/testing environments

S3 Cost Estimation

For PutPlace file storage:

Storage Amount	Monthly Cost (S3 Standard)
1 GB	$0.023
10 GB	$0.23
100 GB	$2.30
1 TB	$23.00

Additional costs:

PUT requests: $0.005 per 1,000 requests
GET requests: $0.0004 per 1,000 requests
Data transfer out: $0.09/GB (first 10 TB/month)

Example: Storing 50GB with 10,000 PUT and 100,000 GET requests per month:

Storage: 50 GB × $0.023 = $1.15
PUT: 10,000 ÷ 1,000 × $0.005 = $0.05
GET: 100,000 ÷ 1,000 × $0.0004 = $0.04
Total: ~$1.24/month

Troubleshooting S3 Storage

Check AWS credentials are configured:

ssh root@165.22.123.45
aws configure list --profile putplace-prod

Test S3 access:

ssh root@165.22.123.45
aws s3 ls --profile putplace-prod
aws s3 ls s3://putplace-prod/ --profile putplace-prod

Check service environment variables:

ssh root@165.22.123.45
systemctl show putplace | grep -E 'AWS|Environment'

View application logs for S3 errors:

ssh root@165.22.123.45
tail -f /var/log/putplace/error.log | grep -i s3

Common S3 errors:

“Unable to locate credentials”
- Check /root/.aws/credentials exists and has correct permissions
- Verify AWS_PROFILE is set in systemd service
- Restart service: systemctl restart putplace
“Access Denied” errors
- Verify IAM user has S3 permissions
- Check bucket name matches IAM policy
- Verify bucket exists: aws s3 ls s3://putplace-prod/
“No such bucket”
- Create bucket: aws s3 mb s3://putplace-prod --region eu-west-1
- Or re-run invoke configure --setup-iam

Monitoring and Backups

Health Checks

# Check application health
curl http://your-droplet-ip/health

# Check service status
ssh root@165.22.123.45 'systemctl is-active putplace'

# Check MongoDB
ssh root@165.22.123.45 'systemctl is-active mongod'

MongoDB Backups

ssh root@165.22.123.45

# Create backup
mongodump --db putplace --out /root/backups/$(date +%Y%m%d)

# Restore backup
mongorestore --db putplace /root/backups/20250113/putplace

Automated Backups

Add to crontab:

ssh root@165.22.123.45
crontab -e

# Daily backup at 3 AM
0 3 * * * mongodump --db putplace --out /root/backups/$(date +\%Y\%m\%d) && find /root/backups -mtime +7 -delete

Digital Ocean Backups

Enable droplet backups in Digital Ocean dashboard:

Settings → Backups → Enable
Cost: 20% of droplet price
Weekly automated snapshots

Scaling

Vertical Scaling (Resize Droplet)

# Power off droplet
doctl compute droplet-action shutdown <droplet-id>

# Resize (this may take several minutes)
doctl compute droplet-action resize <droplet-id> --size s-2vcpu-4gb --resize-disk

# Power on
doctl compute droplet-action power-on <droplet-id>

Horizontal Scaling (Multiple Droplets + Load Balancer)

Create multiple droplets with same app
Set up Digital Ocean Load Balancer
Use MongoDB Atlas (shared database)
Point domain to load balancer

Troubleshooting

Application Won’t Start

# Check service status
systemctl status putplace

# Check logs
journalctl -u putplace -n 50

# Check application logs
tail -50 /var/log/putplace/error.log

# Test manually
cd /opt/putplace/putplace
source .venv/bin/activate
uvicorn putplace.main:app --host 0.0.0.0 --port 8000

MongoDB Connection Issues

# Check MongoDB status
systemctl status mongod

# Check MongoDB logs
tail -50 /var/log/mongodb/mongod.log

# Test MongoDB connection
mongosh --eval "db.adminCommand('ping')"

# Check if MongoDB is listening
netstat -tlnp | grep 27017

Nginx Issues

# Check nginx status
systemctl status nginx

# Test nginx configuration
nginx -t

# Check nginx logs
tail -50 /var/log/nginx/error.log

# Restart nginx
systemctl restart nginx

Out of Memory

# Check memory usage
free -h

# Check top processes
top

# Add swap space (if needed)
fallocate -l 2G /swapfile
chmod 600 /swapfile
mkswap /swapfile
swapon /swapfile
echo '/swapfile none swap sw 0 0' >> /etc/fstab

SSL Certificate Issues

# Check certificate status
certbot certificates

# Renew certificate
certbot renew

# Test auto-renewal
certbot renew --dry-run

Security Best Practices

1. Change SSH Port (Optional)

# Edit SSH config
nano /etc/ssh/sshd_config

# Change port (e.g., to 2222)
Port 2222

# Restart SSH
systemctl restart sshd

# Update firewall
ufw allow 2222/tcp
ufw delete allow 22/tcp

3. Set Up Fail2Ban

apt-get install fail2ban

# Configure
nano /etc/fail2ban/jail.local

# Add:
[sshd]
enabled = true
port = ssh
logpath = /var/log/auth.log
maxretry = 3
bantime = 3600

systemctl restart fail2ban

4. Regular Updates

# Set up automatic security updates
apt-get install unattended-upgrades
dpkg-reconfigure --priority=low unattended-upgrades

Cleanup and Removal

Remove Droplet

# List droplets
doctl compute droplet list

# Delete droplet
doctl compute droplet delete <droplet-id>

# Or by name
doctl compute droplet delete putplace-prod

Remove from Digital Ocean Dashboard

Go to https://cloud.digitalocean.com/droplets
Click on droplet
Click “Destroy” in the menu
Confirm deletion

Advanced Configuration

Custom Systemd Service Settings

Edit /etc/systemd/system/putplace.service:

[Service]
# More workers for high traffic
ExecStart=/opt/putplace/putplace/.venv/bin/uvicorn putplace.main:app --host 0.0.0.0 --port 8000 --workers 8

# Memory limits
MemoryMax=1G
MemoryHigh=800M

# Process limits
LimitNOFILE=65536

Reload and restart:

systemctl daemon-reload
systemctl restart putplace

Nginx Performance Tuning

Edit /etc/nginx/sites-available/putplace:

server {
    listen 80;
    server_name api.example.com;

    # Increase client body size for large uploads
    client_max_body_size 100M;

    # Timeouts
    proxy_connect_timeout 60s;
    proxy_send_timeout 60s;
    proxy_read_timeout 60s;

    # Buffering
    proxy_buffering on;
    proxy_buffer_size 4k;
    proxy_buffers 8 4k;

    # Compression
    gzip on;
    gzip_types application/json;

    location / {
        proxy_pass http://localhost:8000;
        proxy_set_header Host $host;
        proxy_set_header X-Real-IP $remote_addr;
        proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
        proxy_set_header X-Forwarded-Proto $scheme;
    }
}

Using Postgres Instead of MongoDB

The deployment script currently installs MongoDB, but you can modify it to use Postgres:

Edit deployment script to install Postgres instead
Update application to use SQLAlchemy + Postgres
Update connection string in .env

Cost Optimization

Resize Droplet Based on Usage

Start with 1GB ($6/month), monitor for a week, then resize if needed:

# Check memory usage
ssh root@IP 'free -h'

# Check CPU usage
ssh root@IP 'top -bn1 | head -20'

# If consistently using >80% memory, resize to 2GB

Use Monitoring

Enable Digital Ocean monitoring (free):

Dashboard → Droplet → Enable Monitoring
View CPU, memory, disk, and network graphs

Snapshots vs Backups

Backups: Automatic, weekly, 20% of droplet cost
Snapshots: Manual, cheaper for occasional use

Support

Digital Ocean Docs: https://docs.digitalocean.com/
Community: https://www.digitalocean.com/community
Support: https://www.digitalocean.com/support

Comparison with Other Deployment Methods

Feature	Digital Ocean	AWS AppRunner	Heroku
Cost	$6-12/month	$25+/month	$25+/month
Setup Time	~10 minutes	~5 minutes	~5 minutes
Flexibility	Full control	Limited	Limited
Scaling	Manual resize	Auto-scale	Auto-scale
MongoDB	Included	Separate ($9+)	Addon ($15+)
SSL	Free (Let’s Encrypt)	Included	Included
SSH Access	Yes	No	Limited
Custom Software	Yes	No	Limited

Digital Ocean is best for:

Cost-sensitive deployments
Full control requirements
Learning/understanding the stack
Custom software needs

AppRunner/Heroku are best for:

Zero-maintenance deployments
Auto-scaling requirements
Quick prototypes
Teams without DevOps expertise

Digital Ocean Deployment Guide

Overview

Cost Comparison

Quick Start

Prerequisites

Installation

Simplified Deployment (Recommended)

Manual Deployment (Full Control)

Deployment Options

Deploy to Existing Droplet

Deploy with Custom Domain

Custom Droplet Size and Region

Deploy with AWS S3 Storage

Deploy with MongoDB Atlas and S3 Storage

Quick Updates

Management Commands

SSH Access

View Logs

Service Management

Post-Deployment Configuration

1. Create Admin User

2. Enable SSL/HTTPS (Recommended)

3. Configure Firewall (UFW)

4. Configure Environment Variables

Architecture

Server Layout

Services

Network Flow

Using MongoDB Atlas Instead

1. Create Atlas Cluster

2. Update Configuration

Using AWS S3 for File Storage

Prerequisites

1. Generate AWS Credentials (One-Time Setup)

2. Deploy with S3 Storage and AWS Credentials

3. Verify S3 Configuration

4. Alternative: Manual AWS Credentials Setup

5. Environment-Specific Deployments

S3 Storage Benefits

S3 Cost Estimation

Troubleshooting S3 Storage

Monitoring and Backups

Health Checks

MongoDB Backups

Automated Backups

Digital Ocean Backups

Scaling

Vertical Scaling (Resize Droplet)

Horizontal Scaling (Multiple Droplets + Load Balancer)

Troubleshooting

Application Won’t Start

MongoDB Connection Issues

Nginx Issues

Out of Memory

SSL Certificate Issues

Security Best Practices

1. Change SSH Port (Optional)

2. Disable Root SSH Login

3. Set Up Fail2Ban

4. Regular Updates

Cleanup and Removal

Remove Droplet

Remove from Digital Ocean Dashboard

Advanced Configuration

Custom Systemd Service Settings

Nginx Performance Tuning

Using Postgres Instead of MongoDB

Cost Optimization

Resize Droplet Based on Usage

Use Monitoring

Snapshots vs Backups

Support

Comparison with Other Deployment Methods