Skip to main content

Docker

Usage​

# pull image and run stuff through it
docker pull IMAGE_NAME
docker run -v $(pwd):/outputs -it IMAGE_NAME /bin/bash /outputs/build.sh

# ssh into image
docker run -it REPOSITORY

# re-attach container
docker exec -it 3dbcd9237a43 /bin/bash

# list images
docker images

# list containers
docker ps

# list all containers
docker ps -a

# docker stop
docker stop CONTAINER_ID

# remove image
docker rmi node IMAGE ID

# transfer file
docker cp CONTAINER_ID:DOCKER_PATH_TO_ZIP_FILE LOCAL_PATH

# forward port
docker run -it -p ON_HOST_PORT:IN_CONTAINER_PORT baania/pyspark

# prune
docker system prune

# create and attach volume
docker volume create --name hello
docker run -d -v hello:/container/path/for/volume container_image my_command

docker-compose​

version: "3"
services:
  jupyter:
    image: jupyter/pyspark-notebook:latest
    ports:
      - "8888:8888"
      - "4040:4040"
    volumes:
      - .:/home/jovyan
    environment:
      - JUPYTER_ENABLE_LAB=yes

network to host from docker on linux​

   environment:
      NC_DB: "pg://localhost:5432?u=postgres&p=password&d=root_db"
    extra_hosts:
      - "host.docker.internal:host-gateway"
    network_mode: host

Useful snippets​

# remove docker images based on name
docker rmi $(docker images | grep 'imagename')

# use amd64 when you're on arm64
export DOCKER_DEFAULT_PLATFORM=linux/amd64

Resources​

  • composerize - docker run asdlksjfksdf > docker-composerize up.
  • contains.dev - Explore your images, view their files, layers and dependencies.
  • The Compose Specification - The Compose specification establishes a standard for the definition of multi-container platform-agnostic applications.