Fix Kamal deployment OOM crashes with blue-green deployment memory doubling

Root cause: Kamal uses blue-green deployment strategy, which runs TWO instances of your app simultaneously during deployment. Your 75% baseline becomes 150% (100%+), causing OOM kills.

Fix 1 - Minimum 2GB RAM for Kamal:

# Upgrade droplet to 2GB minimum
doctl compute droplet-resize <droplet-id> --size s-1vcpu-2gb

Blue-green deploys inherently require 2x memory headroom.

Fix 2 - Add swap space as temporary buffer:

# SSH into droplet
ssh root@your-droplet-ip

# Create 2GB swap file
fallocate -l 2G /swapfile
chmod 600 /swapfile
mkswap /swapfile
swapon /swapfile

# Make permanent
echo '/swapfile none swap sw 0 0' >> /etc/fstab

Warning: Swap is slow. Use only as emergency overflow, not permanent solution.

Fix 3 - Configure deploy stop_wait_time:

# config/deploy.yml
stop_wait_time: 5  # Reduce from default 30s to free memory faster

Kills old container sooner to reclaim memory.

Fix 4 - Use single-server mode (no blue-green):

# config/deploy.yml
readiness_delay: 0

Warning: This creates downtime during deploys.

Recommended setup:

• Minimum: 2GB RAM for production Kamal deployments
• Optimal: 4GB for multiple small apps on one VM
• Always monitor memory with kamal app logs --tail 100 during deploy