Simplify your dev life with Docker (WebElement Banská Bystrica)

Transcript

1. Simplify your dev life
   with Docker
   Vladimír Kriška (@ujovlado)
   WebElement Banská Bystrica
   

   View Slide

2. Past (vedle pasti)
   ●
   New computer
   ●
   Install Apache|Nginx, MySQL|PostgreSQL, PHP,
   MongoDB, etc., etc.
   ●
   Start services … start coding
   After few weeks
   ●
   OMG this project needs newer version of MySQL
   ●
   Problems
   ●
   And problems
   

   View Slide

3. Future
   ●
   New computer
   ●
   Prepare Dockerfile with exact requirements
   ●
   Build it
   ●
   Run it (with services inside) … start coding
   After few weeks
   ●
   OMG this project needs … whatever
   ●
   Never mind, I'll build a new container
   ●
   Profit
   

   View Slide

4. Yo Dawg, I heard you like computers
   So I put computer into computer,
   So you can computer while you computer
   

   View Slide

5. What is Docker?
   "Docker is an open platform for building,
   shipping and running distributed applications."
   

   View Slide

6. What is Docker for me?
   Primary
   ●
   Engine for working with containers
   ●
   Any virtual machine
   ●
   Testing platform
   ●
   Tool to not install "shit" to your computer
   Not yet
   ●
   Running in production
   

   View Slide

7. What really Docker is?
   Same as your system
   ●
   Set of files with "/sbin/init" file
   So, you can
   ●
   Take parts of your system and virtualize them
   without need to virtualize full OS
   

   View Slide

8. Containers without Docker (!)
   https://chimeracoder.github.io/docker-without-docker
   mkdir debian-tree
   debootstrap --arch=amd64 unstable debian-tree
   systemd-nspawn -D debian-tree/ /bin/bash
   machinectl list
   machinectl poweroff debian-tree
   

   View Slide

9. OK, let's go
   Three basic steps
   1) Create Dockerfile
   2) Build it
   3) Run it
   

   View Slide

10. Dockerfile & Build & Run
    FROM debian:jessie
    RUN apt-get update && apt-get upgrade -y
    RUN apt-get install nginx -y
    RUN mkdir -p /data
    COPY package.tar.gz /data/
    EXPOSE 80
    docker build -t wblmnt/bb .
    docker run -i -t --rm -p 8037:80 wblmnt/bb:latest
    

    View Slide

11. Case I & Case II
    Everything in one container
    ●
    Easy to start
    ●
    Same as any virtual machine / container
    ●
    Need to commit changes at the end (as many
    times as containers count)
    

    View Slide

12. Case III
    Multiple containers with separate store
    ●
    For production environment
    ●
    Hard to setup
    ●
    Deployment?
    

    View Slide

13. Examples
    

    View Slide

14. Thanks
    

    View Slide