How to Update Docker Images Without Downtime

The Problem

Running docker compose down && docker compose up causes downtime. For production services, you need a strategy that maintains availability during updates.

Method 1: Docker Compose Rolling Update

# Pull new images
docker compose pull

# Recreate only changed services
docker compose up -d

Docker Compose recreates containers one at a time when using up -d after pulling new images.

Method 2: Blue-Green with Compose Profiles

services:
  app-blue:
    image: myapp:v1
    profiles: ["blue"]
    ports:
      - "3001:3000"
  app-green:
    image: myapp:v2
    profiles: ["green"]
    ports:
      - "3002:3000"

# Start new version
docker compose --profile green up -d

# Test green
curl http://localhost:3002/health

# Switch Nginx upstream to green
# Stop old version
docker compose --profile blue down

Method 3: Docker Swarm Rolling Update

docker service update --image myapp:v2 --update-parallelism 1 --update-delay 10s myapp

Health Check Integration

services:
  app:
    image: myapp:latest
    healthcheck:
      test: ["CMD", "curl", "-f", "http://localhost:3000/health"]
      interval: 10s
      timeout: 5s
      retries: 3
      start_period: 30s
    deploy:
      update_config:
        order: start-first  # Start new before stopping old

Automated Update Script

#!/bin/bash
cd /opt/myapp
docker compose pull
docker compose up -d --remove-orphans
docker image prune -f
echo "Updated at $(date)" >> /var/log/docker-updates.log