Skip to main content

Docker Stacks

Docker Stacks is a higher-level concept used for deploying multi-container applications in Docker Swarm mode. A stack allows you to define and manage the deployment of your services in a distributed, clustered environment with Docker Swarm.

This guide provides an overview of Docker Stacks, their components, and how to use them effectively.

Table of Contents​

What is Docker Stack?​

A Docker Stack is a collection of services that can be deployed to a Docker Swarm cluster. It allows you to define multiple services that should be deployed together, with their configurations and relationships, in a single file. Docker Stacks are similar to Docker Compose files but are specifically designed for use in a Docker Swarm environment.

Stacks are defined in a docker-stack.yml file, which is similar to the docker-compose.yml file, but it includes the ability to deploy and manage services across multiple nodes in a Swarm cluster.

Docker Stack Components​

A typical Docker Stack consists of the following components:

  1. Services: Each service is a container or a set of containers that perform a specific function, such as a web server, a database, etc.
  2. Networks: Services within a stack communicate through Docker networks.
  3. Volumes: Volumes are used for data persistence, allowing services to store data outside the container's filesystem.
  4. Secrets: For managing sensitive data such as passwords or API keys.

Each of these components can be defined in the docker-stack.yml file.

Creating a Docker Stack​

To create a Docker Stack, you need a docker-stack.yml file. This file contains the definitions for the services, networks, and volumes that make up the stack.

Example docker-stack.yml file:​

version: "3.8"

services:
  web:
    image: nginx:latest
    deploy:
      replicas: 3
      resources:
        limits:
          cpus: "0.5"
          memory: 50M
    networks:
      - frontend

  db:
    image: postgres:latest
    environment:
      POSTGRES_PASSWORD: example
    volumes:
      - db-data:/var/lib/postgresql/data
    networks:
      - backend

volumes:
  db-data:

networks:
  frontend:
  backend:

The web service uses the nginx image and runs 3 replicas for load balancing. The db service uses the postgres image and sets an environment variable for the password. The services are connected to different networks: frontend for web and backend for db.

Deploying a Docker Stack​

Once you have created your docker-stack.yml file, you can deploy the stack using the docker stack deploy command.

Command to deploy the stack:​

docker stack deploy -c docker-stack.yml mystack

In this command:

  • -c docker-stack.yml specifies the path to your stack definition file.
  • mystack is the name of the stack.

Docker Swarm will then deploy all the services, networks, and volumes as defined in the stack file to the cluster.

Managing a Docker Stack​

After deploying a stack, you can manage it with several commands:

View stack services:​

docker stack services mystack

View stack containers:​

docker stack ps mystack

Remove a stack:​

docker stack rm mystack

This command will stop and remove all services in the stack.

Scaling Docker Stacks​

One of the key features of Docker Stacks is the ability to scale services. This can be done by changing the number of replicas for a service in the docker-stack.yml file or by using the docker service scale command.

Example: Scaling a service​

To scale the web service to 5 replicas:

To scale the web service to 5 replicas:

You can also adjust the number of replicas by modifying the deploy.replicas value in the docker-stack.yml file and then redeploying the stack.

Best Practices for Docker Stacks​

Here are some best practices for working with Docker Stacks:

  1. Use Named Volumes for Persistence: Always use named volumes for data persistence in Docker Swarm. This ensures data is not lost when containers are removed or moved between nodes.
  2. Use Secrets for Sensitive Data: Instead of embedding sensitive information in environment variables, use Docker Secrets for managing passwords and API keys securely.
  3. Keep Services Stateless: Make services stateless so that they can be easily replicated and moved across nodes in the Swarm cluster.
  4. Leverage Docker Swarm's Auto-healing: Docker Swarm automatically restarts failed containers, so ensure services are configured to be resilient to failures.
  5. Use Overlay Networks: Use Docker Swarm's overlay networks to allow containers to communicate securely across different nodes.

Conclusion​

Docker Stacks provide a powerful way to deploy and manage multi-container applications in a Docker Swarm environment. With the ability to define services, networks, volumes, and secrets in a single YAML file, Docker Stacks make it easier to deploy, scale, and manage applications in a distributed environment.

For more details, check out the official Docker Stacks documentation..