Published on

Docker

Authors
  • avatar
    Name
    Rammy
    Twitter

Docker Basics

/static/images/Others/docker_components.png
/static/images/Others/docker_engine_2.png
command info
docker version
docker -v
docker info
docker --help
docker login

Images

  • Each docker image is a layer of stacked mini docker images, such that when an update happens only those required layers will update.
  • Docker image is created from docker file.
command info
docker images show all images
docker run image_name pull image and create container
docker pull image_name
docker images -q Show image ids
docker images -f "dangling=false" filter images --> show images not associated with any container
docker rmi image_name remove image
docker rmi $( docker images -q ) remove all images not attached to container

Image --> Containers

  • Docker image is like a set of instructions.
  • To actually bring the image alive we create containers from that image.
Command Info
docker ps show running containers
docker ps -a show all containers
docker run --name container_name -it image_name /bin/bash run container in interactive mode
docker start + stop
docker pause + unpause + kill
docker stats container_id/name provide container memory, io usage
docker top container_id/name display running process of container
docker container rm container_id delete container
docker exec -it container_id bin/bash interact with container by bash
docker logs container_id see docker logs
docker container inspect container_id get all details of the container (port, mount,network etc)

Docker File --> Create Image

/static/images/Others/docker_file.png
  • Create Image from Dockerfile : docker build -t myimageName:tag pathToDockerFile
Dockerfile Instructions
FROM node Base image to start from, it could be from scrath as well by using scratch repository from hub
MAINTAINER rammy pal kekagmail.com Maintainer of the dockerfile
RUN mkdir -p /home/app RUN executes with in the container when image is created
COPY ./app /home/app COPY executes on host machine, basically to copy project files to container destination
CMD ["node","/home/app/server.js"] CMD executes when container is created, here when container is started we are running node application
-
/static/images/Others/docker_file_instructions.png

Push Image --> Repository

  • In the above step we created our own image from a docker file. Now we can push this local image to a repository.
  • The repository could be a private or public repository. Lets use dockerhub repository for simplicity.
Docker push steps Explanation
docker pull hello-world First lets pull a simple image -> hello-world:latest from docker hub
docker tag hello-world:latest docker.io/rammyram/myhello:1.0 Make sure image is renamed in the following format /repository_address/image_name:tag
docker push docker.io/rammyram/myhello:1.0 Now you can safely push to your repo in docker hub, of course before that you need to do docker login
docker pull docker.io/rammyram/myhello Pulling your repository

Docker Network

  • TODO

Docker Compose

  • Docker-compose would be used on the server to deploy all the applications/services.
  • The idea is to just provide a single docker compose file to development or testing or deployment team to build the complete project
  • Why we need docker compose :
    • Consider the case of running mongodb and mongo express client containers on a docker network, the below are the steps needed.
Steps Def
docker network create create a network which is shared by both containers
docker run -d -p 27017:27017 -e MONGO_INITDB_ROOT_USERNAME=admin -e MONGO_INITDB_ROOT_PASSWORD=password123 --net myapp --name mongodb mongo start mongodb database container on network myapp
docker run -d -p 8081:8081 -e ME_CONFIG_MONGODB_ADMINUSERNAME=admin -e ME_CONFIG_MONGODB_ADMINPASSWORD=password123 -e ME_CONFIG_MONGODB_SERVER=mongodb --network myapp --name mongo-express mongo-express start mongo express web client container on network myapp and port mapping on 8081

  • What if we have 10 containers or more, we can't keep running these long commands all the time. So we come up with docker compose where we can compose all these commands in a single file.

  • Example :

    • Lets consider the case where our project has 3 services

      • database : mongodb
      • db client : mongo-express
      • shopee_myapp : my nodejs project

    • Step 1 : Create images for all services :

      • mongo : ublic image available from docker hub
      • mongo-express : public image available from docker hub
      • shoppee_myapp: need to create image for our project
        1. create docker file first
        2. create image from docker file
        3. push image to private or public repository

    • Step 2 : Create docker-compose file will all configurations connecting all the 3 services. Check example below -->

version: '3'

services:

 mongodb:
 
 image: mongo
 ports: 
 - 27017:27017
 environment: 
 - MONGO_INITDB_ROOT_USERNAME=admin
 - MONGO_INITDB_ROOT_PASSWORD=password123

 mongo-express:
 
 image: mongo-express
 restart: always
 ports: 
 - 8081:8081
 environment: 
 - ME_CONFIG_MONGODB_ADMINUSERNAME=admin 
 - ME_CONFIG_MONGODB_ADMINPASSWORD=password123 
 - ME_CONFIG_MONGODB_SERVER=mongodb
 depends_on: 
 - mongodb
 
 
 shoppee_myapp: 
 
 image: docker.io/rammyram/myapp:1.2
 ports:
 - 3000:3000
 environment: 
 - MONGO_DB_USERNAME=admin
 - MONO_DB_PWD=password123
 depends_on:
 - mongodb

  • Commands :

commands info
Tool for defining & running multi-container docker applications
use yaml files to configure application services (docker-compose.yml)
docker-compose -f docker-compose.yaml up start all services in docker-compose with a single command
docker-compose -f docker-compose.yaml down down all services -> down containers + remove containers
docker-compose config check any errors in compose yaml file
docker-compose -d up in detached mode
docker-compose up --scale mongodb=4 scaling a services

Volumes

/static/images/Others/docker_volume.png
  • Decoupling container from storage
  • Share volume (storage/data) among different containers
  • Attach volume to container
  • On deleting container data still persists in volumes.

  • Types of volumes :

Command Type of volume
docker run -v /home/mount/data:/var/lib/mysql/data Host volumes -> you decide where on the host the reference is made to save data
docker run -v /var/lib/mysql/data Anonymous volumes -> docker takes care of automatically mounting to a location on host ( ex: /var/lib/docker/randon-hash/data)
docker run -v app_vol:/var/lib/mysql/data Named Volumes --> we can create a specific docker volume and set to use that volume. These named volumes are located on host ( ex: /var/lib/docker/volumes/app_vol/data)...suitable for production and also simpler when volume needs to be shared among containers

Stats

Command Stats
docker stats Details on running containers, memory usage
docker system df Disk usage of docker, all images, containers, local volumes, size
docker system prune - Delete stopped containers - Release any unused network & memory

Status: #done

Tags: #docker_basics #docker_image #docker_container #docker_image #docker_compose

References:

Related:

Discuss on Twitter • View on GitHub