Creating docker custom image

Creating docker custom image

• 1.
  정찬민 Docker Custom Image
• 2.
  목차 • Docker • VM과의 차이 • Docker Image • command • Make Custom Image
• 3.
  Docker • 컨테이너 기반의오픈 소스 가상화 플랫폼 • 애플리케이션을 신속하게 구축, 테스트 및 배 포 가능 • 컨테이너에는 라이브러리, 시스템 도구, 코드, 런타임 등 소프트웨어를 실행하는 데 필요한 모든 것이 포함 • 환경에 구애 받지 않고 애플리케이션을 신속 하게 배포 및 확장
• 4.
  VM 과의 차이 •OS+APP+LIB 로 구성된 VM 과는 달리 Docker 는 APP+LIB 만 설치 • OS 자원은 Host OS 와 공유 [VM] [Docker] 가상 머신에 비해 경량화가 가능하며 빠른 처리 속도
• 5.
  Docker Image [Docker hub][Docker host] image pull • Docker 컨테이너를 만들기 위한 템플릿 • 이미지는 컨테이너 실행에 필요한 파일 시스템, 라이브러리, 설정 정보등을 포함 • 같은 이미지에서 여러 개의 컨테이너를 실행할 수 있고 컨테이너가 변해도 이미지 고정
• 6.
  Command • docker run– docker image 로 컨테이너 생성 docker run -d -it --name <docker image 이름> < docker hub reference image> • docker exec – 로컬에서 docker 컨테이너 안의 명령을 실행 docker exec -it <CONTAINER ID> <shell> • docker tag – docker hub 에 이미지를 업로드 하기 위해서는 tag 설정 docker tag <IMAGE> <저장소>/<image 이름>:<버전> • docker push – docker image 를 docker hub 에 업로드 docker push <저장소>/<image 이름>:<버전> docker push <docker hub 계정>/<image 이름>:<버전> • docker commit – 컨테이너 생성 이후 변경 사항을 docker image 에 업로드 docker commit <CONTAINER ID> <image 이름>:<버전>
• 7.
  Command • docker cp– docker 컨테이너와 로컬과의 파일 복사 docker cp <로컬 파일 명> <CONTAINER ID>:<저장 위치> docker cp <CONTAINER ID>:<저장 위치> <로컬 파일 명>
• 8.
  Make Custom Image •APP : python 3.8 – Miniconda3 • APP : java8.0 - openjdk • 고객사 APP 실행 시 필요 library 설치 ubuntu 18.04 + miniconda3 + library ubuntu 18.04 + miniconda ubuntu 18.04 + miniconda + library Dockertest:0.1 Dockertest:0.2 Dockertest:0.3 + jdk
• 9.
  Make Custom Image •docker hub 를 통해 ubuntu18.04 image 로 docker 컨테이너 생성 docker run -d -it --name dockertest ubuntu:18.04 root@bizz:/home/bizz# docker run -d -it --name dockertest ubuntu:18.04 Unable to find image 'ubuntu:18.04' locally 18.04: Pulling from library/ubuntu f08d8e2a3ba1: Pull complete 3baa9cb2483b: Pull complete 94e5ff4c0b15: Pull complete 1860925334f9: Pull complete Digest: sha256:05a58ded9a2c792598e8f4aa8ffe300318eac6f294bf4f49a7abae7544918592 Status: Downloaded newer image for ubuntu:18.04 c9a417d374a292db4caebe55abd890986e5a6ca8fc4aba301c8ca058cce8e154 root@bizz:/home/bizz# ls miniconda3 miniconda3_edit.tar.gz miniconda3.tar.gz mount snap root@bizz:/home/bizz# docker ps CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES c9a417d374a2 ubuntu:18.04 "/bin/bash" 56 seconds ago Up 52 seconds dockertest
• 10.
  Make Custom Image •ubuntu 에 miniconda3 설치 wget https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh bash Miniconda3-latest-Linux-x86_64.sh root@c9a417d374a2:/# wget https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh --2020-08-29 07:26:10-- https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh Resolving repo.anaconda.com (repo.anaconda.com)... 104.16.130.3, 104.16.131.3, 2606:4700::6810:8303, ... Connecting to repo.anaconda.com (repo.anaconda.com)|104.16.130.3|:443... connected. HTTP request sent, awaiting response... 200 OK Length: 93052469 (89M) [application/x-sh] Saving to: 'Miniconda3-latest-Linux-x86_64.sh' Miniconda3-latest-Linux-x86_64.sh 100%[=================================================================================================== =====================================>] 88.74M 10.9MB/s in 9.8s 2020-08-29 07:26:19 (9.04 MB/s) - 'Miniconda3-latest-Linux-x86_64.sh' saved [93052469/93052469] root@c9a417d374a2:/# bash Miniconda3-latest-Linux-x86_64.sh <생략> Thank you for installing Miniconda3! root@c9a417d374a2:/# • docker 컨테이너 접속 docker exec –it c9a417d374a2 bash root@bizz:/home/bizz# docker exec -it c9a417d374a2 bash root@c9a417d374a2:/#
• 11.
  Make Custom Image •1차 이미지 배포 docker commit c9a417d374a2 docker_custom_image:0.1 docker tag docker_custom_image:0.1 jcm9812/docker_custom_image:0.1 docker push jcm9812/docker_custom_image:0.1 root@bizz:/home/bizz# docker ps CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES c9a417d374a2 ubuntu:18.04 "/bin/bash" 26 minutes ago Up 26 minutes dockertest root@bizz:/home/bizz# docker commit c9a417d374a2 docker_custom_image:0.1 sha256:61362b1e924abe4b1384503ce503a9956b3a915e06a66a247e79eff11e4cce25 root@bizz:/home/bizz# docker tag docker_custom_image:0.1 jcm9812/docker_custom_image:0.1 root@bizz:/home/bizz# docker push jcm9812/docker_custom_image:0.1 The push refers to repository [docker.io/jcm9812/docker_custom_image] e335ccd7dfd0: Pushed 001e4a80973b: Layer already exists 2ba5b91ca2b0: Layer already exists 2f37d1102187: Layer already exists 79bde4d54386: Layer already exists 0.1: digest: sha256:f90857818af46db53efbb9ea312dbdd169984db7a14432cbeee3082fd942e1a9 size: 1365 root@bizz:/home/bizz#
• 12.
  Make Custom Image •1차 배포 이미지 로 docker 컨테이너 생성 docker run -d -it jcm9812/docker_custom_image:0.1 - local 에 docker image 가 없을 경우 docker run -d -it docker_custom_image:0.1 - local 에 docker image 가 있을 경우 root@bizz:/home/bizz# docker run -d -it jcm9812/docker_custom_image:0.1 010fecc8215a38ce3cedca84388e2ac4948e609777bb6d32d5d1f15ea7cd0105 root@bizz:/home/bizz# docker ps CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES 010fecc8215a jcm9812/docker_custom_image:0.1 "/bin/bash" 4 seconds ago Up 2 seconds great_ramanujan c9a417d374a2 ubuntu:18.04 "/bin/bash" 2 hours ago Up 2 hours dockertest root@bizz:/tmp# docker cp numpy-1.19.1-cp38-cp38-manylinux1_x86_64.whl 010fecc8215a:/tmp root@bizz:/tmp# docker ps CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES 010fecc8215a jcm9812/docker_custom_image:0.1 "/bin/bash" 53 minutes ago Up 53 minutes great_ramanujan c9a417d374a2 ubuntu:18.04 "/bin/bash" 2 hours ago Up 2 hours root@bizz:/tmp# docker exec -it 010fecc8215a bash (base) root@010fecc8215a:/# ls /tmp/ numpy-1.19.1-cp38-cp38-manylinux1_x86_64.whl (base) root@010fecc8215a:/# • 라이브러리 컨테이너에 복사(ex : numpy-1.19.1 ) docker cp numpy-1.19.1-cp38-cp38-manylinux1_x86_64.whl 010fecc8215a:/tmp docker exec –it 010fecc8215a bash
• 13.
  Make Custom Image •필요 라이브 러리 및 패키지 설치 /root/miniconda3/bin/python -m pip install /tmp/ tmp/numpy-1.19.1-cp38-cp38-manylinux1_x86_64.whl (base) root@010fecc8215a:/# /root/miniconda3/bin/python -m pip install /tmp/numpy-1.19.1-cp38-cp38-manylinux1_x86_64.whl Processing /tmp/numpy-1.19.1-cp38-cp38-manylinux1_x86_64.whl Installing collected packages: numpy Successfully installed numpy-1.19.1 (base) root@010fecc8215a:/#
• 14.
  Make Custom Image •2차 이미지 빌드 docker commit 010fecc8215a docker_custom_image:0.2 docker tag docker_custom_image:0.1 jcm9812/docker_custom_image:0.2 docker push jcm9812/docker_custom_image:0.2 root@bizz:/tmp# docker ps CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES 010fecc8215a jcm9812/docker_custom_image:0.1 "/bin/bash" About an hour ago Up About an hour great_ramanujan c9a417d374a2 ubuntu:18.04 "/bin/bash" 3 hours ago Up 3 hours dockertest root@bizz:/tmp# docker commit 010fecc8215a docker_custom_image:0.2 sha256:ad641601e269ac243db4f634fa7c3327e4c1e4d1377870ec574b7194ec141756 root@bizz:/tmp# docker tag docker_custom_image:0.2 jcm9812/docker_custom_image:0.2 root@bizz:/tmp# docker push jcm9812/docker_custom_image:0.2 The push refers to repository [docker.io/jcm9812/docker_custom_image] 897b94cf4ef6: Pushed e335ccd7dfd0: Layer already exists 001e4a80973b: Layer already exists 2ba5b91ca2b0: Layer already exists 2f37d1102187: Layer already exists 79bde4d54386: Layer already exists 0.2: digest: sha256:f1da086d73933235864ae1134290dcfdac69dd8faa049426251b5ba154f47319 size: 1577
• 15.
  Make Custom Image •최종 이미지 배포(Dockerfile) - Docker image 생성을 위한 스크립트 # docker image load FROM jcm9812/docker_custom_image:0.2 # docker image load COPY --from=openjdk:8-alpine / / # 내부 명령어 실행 RUN /root/miniconda3/bin/python -version # home directory 설정 WORKDIR /home/bizz # 내부 명령어 실행 RUN ["/bin/bash", "-c", "java -version"] Dockerfile ▪ Docker image 생성을 위한 스크립트 파일 ▪ Docker exec / 설치 / commit 의 일괄 작업의 스크립팅 ▪ docker hub 로 부터 1개 이상의 docker image 사용 가능
• 16.
  Make Custom Image •최종 이미지 배포 root@bizz:/tmp# docker build -t jcm9812/docker_custom_image . Sending build context to Docker daemon 13.35MB Step 1/5 : FROM jcm9812/docker_custom_image:0.2 0.2: Pulling from jcm9812/docker_custom_image Digest: sha256:f1da086d73933235864ae1134290dcfdac69dd8faa049426251b5ba154f47319 Status: Downloaded newer image for jcm9812/docker_custom_image:0.2 ---> ad641601e269 Step 2/5 : COPY --from=openjdk:8-alpine / / ---> Using cache ---> daffe858dbdd Step 3/5 : RUN /root/miniconda3/bin/python --version ---> Running in ebfecf1ae0b4 Python 3.8.3 Removing intermediate container ebfecf1ae0b4 ---> 38905c06c230 Step 4/5 : WORKDIR /home/bizz ---> Running in 4a72d8a73347 Removing intermediate container 4a72d8a73347 ---> 7a0ab5136f38 Step 5/5 : RUN ["/bin/bash", "-c", "java -version"] ---> Running in ebf8688d442b openjdk version "1.8.0_212" OpenJDK Runtime Environment (IcedTea 3.12.0) (Alpine 8.212.04-r0) OpenJDK 64-Bit Server VM (build 25.212-b04, mixed mode) Removing intermediate container ebf8688d442b ---> 95b5b0161d7b Successfully built 95b5b0161d7b Successfully tagged jcm9812/docker_custom_image:latest root@bizz:/tmp# docker image ls REPOSITORY TAG IMAGE ID CREATED SIZE jcm9812/docker_custom_image latest 95b5b0161d7b 5 seconds ago 737MB jcm9812/docker_custom_image 0.2 ad641601e269 14 minutes ago 632MB docker_custom_image 0.2 ad641601e269 14 minutes ago 632MB jcm9812/docker_custom_image 0.1 61362b1e924a 3 hours ago 524MB docker_custom_image 0.1 61362b1e924a 3 hours ago 524MB jcm9812/dockertest2 0.2 be612743ab2c 7 days ago 70MB dockertest2 0.2 be612743ab2c 7 days ago 70MB root@bizz:/tmp#
• 17.
  Make Custom Image •최종 이미지 확인 root@bizz:/tmp# docker run -d -it jcm9812/docker_custom_image ef699044e1d1b2d7936994cd6b9c9b5131ee89757a11c436b30ffac5aeddad75 root@bizz:/tmp# docker ps CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES ef699044e1d1 jcm9812/docker_custom_image "/bin/bash" 3 seconds ago Up 1 second objective_gagarin 010fecc8215a jcm9812/docker_custom_image:0.1 "/bin/bash" 2 hours ago Up 2 hours great_ramanujan c9a417d374a2 ubuntu:18.04 "/bin/bash" 3 hours ago Up 3 hours dockertest root@bizz:/tmp# docker exec -it ef699044e1d1 bash (base) root@ef699044e1d1:/home/bizz# java -version openjdk version "1.8.0_212" OpenJDK Runtime Environment (IcedTea 3.12.0) (Alpine 8.212.04-r0) OpenJDK 64-Bit Server VM (build 25.212-b04, mixed mode) (base) root@ef699044e1d1:/home/bizz# /root/miniconda3/bin/python --version Python 3.8.3 (base) root@ef699044e1d1:/home/bizz#
• 18.
  감사합니다