5 minutes to learn about docker

 1. Concept:
 
An open source application container engine that packages developers ’applications and dependencies in a portable container, equivalent to a sandbox.
 
Docker allows developers to isolate and run multiple applications on a single operating system instead of dedicating a virtual machine to each application on the server.
 
This is achieved by isolating applications in separate containers. Although these applications are separated by the container, they can share the operating system and other resources.
 
Advantages: The use of containers is more lightweight, can reduce costs, better use of resources and play a higher performance.
 
2. Composition:
 
Docker is mainly composed of clients, daemons, images, and containers.
 
Three. Dockerfile writing
 
FROM python: v3. 6 #Load base image
maintainer name, mail # created, name, mailbox
COPY. /Requirements.txt / root / requirements.txt # Copy local files to the mirror
WORKDIR / root # Configure the path
CMD [ " / bin / python " , " /app.py " , " run " ] #Start the main program
 
 
Four, common docker commands
Find mirror
docker search ouruser/sinatra

Pull image to local
docker pull ouruser/sinatra:v2

Create mirror
docker build -t ouruser/sinatra:v2 .

The specified port is started by default
docker run -d -p 10022:22 -p 5000:5000 9237f60b18d0

Specify the port bash to start
docker run -t -i -p 10022:22 -p 5000:5000 9237f60b18d0 /bin/bash

Specify the container name to start
docker run -it -d  --name c847084bed94 -p 5001:5001 -p 5055:5055 zhongzh/demo

Automatic restart, the maximum number is 10
docker run --restart=on-failure:10 redis

Move files inside and outside the container
docker cp filename zhongzh@host_ip:/home/

Enter container view
sudo docker exec -it 3c6b44cbcc8f /bin/bash

Stop / start container
docker stop/start 3c6b44cbcc8f

View mirror
docker images

View currently running containers
docker ps

Submit Modified Container
docker commit 9237f60b18d0 demo:v1.0
 
Fifth, the actual experience of docker
1. Simplified the environment configuration of deep learning
2. Various deployment environments are based on fools
3. Reusable and portable
 
 

Guess you like

Origin www.cnblogs.com/zhongzihao/p/12727453.html