Docker

Running Docker Compose in production

Published 2026–04–08 • 10 min read

Docker Compose is excellent for local development. Running it in production requires a few additional considerations: resource limits, health checks, log management, and a strategy for zero-downtime updates.

Resource limits

Always set mem_limit and cpus on every service. Without limits, a misbehaving container can consume all available memory and crash the host.

services:
  app:
    image: myapp:latest
    mem_limit: 512m
    cpus: 0.5
    restart: unless-stopped

Health checks

Define a healthcheck so Docker knows when a container is actually ready to serve traffic — not just started.

healthcheck:
  test: ["CMD", "curl", "-f", "http://localhost:3000/health"]
  interval: 30s
  timeout: 5s
  retries: 3
  start_period: 10s

Logging

By default Docker uses the json-file driver with no rotation. Set a max size to prevent logs from filling your disk.

logging:
  driver: json-file
  options:
    max-size: "10m"
    max-file: "3"

Zero-downtime updates

Docker Compose doesn’t support rolling updates natively. The simplest approach: use docker compose up -d --no-deps service to restart a single service. For true zero-downtime, run two instances behind a reverse proxy (Nginx Front Door) and update them one at a time.

Persistent data

Never store persistent data inside containers. Always use named volumes or bind mounts to a path on persistent block storage. This ensures data survives container rebuilds and host reboots.

Docker hosting on Simplewala