Health checks in Compose in Docker - Commands & Configuration

Introduction

Sometimes a service inside a container might start but not be ready to work. Health checks help Docker Compose know if a service is healthy and ready to use by running simple tests inside the container.

When you want to make sure a web server inside a container is fully ready before other services connect to it

When you run a database container and want to check if it is accepting connections

When you have multiple containers and want to restart a container automatically if it becomes unhealthy

When you want to delay starting dependent services until the main service is healthy

When you want to monitor the status of your containers easily with Docker Compose commands

Config File - docker-compose.yml

docker-compose.yml

version: '3.8'
services:
  web:
    image: nginx:1.23
    ports:
      - "8080:80"
    healthcheck:
      test: ["CMD", "curl", "-f", "http://localhost"]
      interval: 30s
      timeout: 10s
      retries: 3
      start_period: 5s

This Compose file defines a service named web using the nginx image.

The healthcheck section runs a command inside the container to check if the web server is responding.

• test: The command to run for the health check. Here it tries to fetch the homepage using curl.
• interval: How often to run the check (every 30 seconds).
• timeout: How long to wait for the check to finish (10 seconds).
• retries: How many times to retry before marking unhealthy (3 times).
• start_period: Time to wait after container start before checking (5 seconds).

Commands

Starts the services defined in the docker-compose.yml file in detached mode so they run in the background.

Terminal

docker-compose up -d

Expected Output

Creating network "default" with the default driver

Creating default_web_1 ... done

→

-d - Run containers in the background (detached mode)

Shows the status of the running containers including their health status.

Terminal

docker-compose ps

Expected Output

    Name                  Command               State                 Ports               

--------------------------------------------------------------------------------

default_web_1   nginx -g 'daemon off;'   Up (healthy)   0.0.0.0:8080->80/tcp

Shows detailed health check results for the web container in JSON format.

Terminal

docker inspect --format='{{json .State.Health}}' default_web_1

Expected Output

{"Status":"healthy","FailingStreak":0,"Log":[{"Start":"2024-06-01T12:00:00Z","End":"2024-06-01T12:00:05Z","ExitCode":0,"Output":""}]}

Stops and removes the containers, networks, and volumes created by docker-compose up.

Terminal

docker-compose down

Expected Output

Stopping default_web_1 ... done

Removing default_web_1 ... done

Removing network default

Key Concept

If you remember nothing else from this pattern, remember: health checks let Docker Compose know if a container is ready and healthy by running simple commands inside it.

Common Mistakes

Not specifying a health check command or using a command that always succeeds

Docker will not be able to detect if the service is actually ready or healthy, so it may report unhealthy containers as healthy.

Use a command that truly tests the service readiness, like a curl request to a web server or a database connection test.

Setting too short intervals or timeouts for health checks

The container might be marked unhealthy too quickly during normal startup delays, causing unnecessary restarts.

Set reasonable intervals, timeouts, and start_period to allow the service enough time to start.

Ignoring health status in dependent services

Other services might start before the main service is healthy, causing failures or errors.

Use health check status to control service startup order or retries.

Summary

Define health checks in docker-compose.yml to test if a container service is ready.

Use docker-compose up to start services and docker-compose ps to see health status.

Inspect detailed health results with docker inspect and clean up with docker-compose down.