Introduction to docker
docker pull website
Docker - Official Image | Docker Hub
1. Install docker
1. Delete the local docker
yum remove docker*
2. Download the yum runtime library
sudo yum install -y yum-utils
3. Set download source
yum-config-manager --add-repo https://download.docker.com/linux/centos/docker-ce.repo
4. Download docker
sudo yum install -y docker-ce docker-ce-cli containerd.io
-The Query options of the y parameter are set to yes, and the docker server, docker command line and dynamic link files used by docker are installed respectively
5. Set startup
systemctl enable docker --now
6. Set the image accelerator
Find mirror acceleration source
Alicloud server - > container mirroring Service - > mirroring tool - > mirroring accelerator - > select centos
sudo mkdir -p /etc/docker
sudo tee /etc/docker/daemon.json <<-'EOF'
{
"registry-mirrors": ["https://6nm7p6th.mirror.aliyuncs.com"]
}
EOF
sudo systemctl daemon-reload
sudo systemctl restart docker
Find image
hub.docker.com
Download Image
docker pull Software name
Common commands
docker images #View the downloaded images docker ps #View the software started and running on the docker server docker rmi image Id #Remove mirror docker logs image id #View the log of the mirror docker cp Started mirror id: /etc/nginx/nginx.conf /data/conf/nginx.conf #Copy the things in the designated position of the container, or copy the things outside to the inside of the container
Start container
docker run Set item image name #Mirror start run command docker run --name=mynginx -d nginx --name Give a name to the started container to distinguish it -d Let the container run in the background docker rm id/Container name #Kill process - f this parameter kills the running process. First, use docker ps -a to view the suspended containers that have been terminated docker stop id/Container name #Stop the container started in the background running mode. It just stops and still exists in the process docker start id/Container name #Start the stopped container
After the running machine restarts, the running container stops. You can add -- restart=always when starting. After the machine restarts, the container starts automatically
-p port mapping 88:80 forwards the access request received by port 88 to port 80
Enter the container to modify the content
(example: change the html file of nginx)
1. Use the docker ps command to query the id of the currently running container
2. Execute docker exec - it < container ID > / bin/bash to switch to the virtual machine operating on the container (/ bin/bash refers to the current command window. If / bin/bash cannot be used, try / bin/sh)
3. Operate the container
4.exit exit docker container
Submit changes
(saving the changes to the container is equivalent to saving a snapshot, similar to git commit)
docker commit [OPTIONS] CONTAINER [REPOSITORY[:TAG]]
1. Execute the command
docker commit [option] <container id> <Alias the modified container>
Options:
-a, --author string Author (e.g., "John Hannibal ")
-c, --change list Apply Dockerfile instruction to the created image
-m, --message string Commit message
-p, --pause Pause container during commit (default true)
2. Example: docker commit -a "wood" - m "home page change" < container ID > wood nginx: v1.0
3.docker rm container id # close the started container
4. Use docker images to see the modified container
5. Start the modified container with docker run - D wood nginx: v1.0
Run the modified container on the new machine
Method 1: compress the image and generate a compressed package to save the image
1. Image name submitted by docker save - O xxx.tar: version tag
docker save -o wood-nginx.tar wood-nginx:v1.0
After the command is executed, the image compressed file is generated in the file directory where the command is executed
2. Transfer to other machines and start the image after decompression
docker load -i Compressed file name of the image docker load -i wood-nginx.tar
3. After the target machine executes docker images, you can see the decompressed image name
Method 2: push the image to the remote warehouse
1. Register an account on hub.docker.com and establish a public warehouse with your own authority
2. The image name modified by docker tag: the absolute path of version tag in the warehouse: version tag (changed to the name format required by the warehouse)
docker tag wood-nginx:v1.0 wood/wood-nginx:v1.0
3. Using docker images, you can see an additional image name with the path as the name
4. Log in to the remote warehouse at the terminal
docker login
Enter the user and password registered on the website
5. Push the image with path prefix to the remote warehouse
docker push wood/wood-nginx:v1.0
6. Log out
docker logout
7. View the image pushed on the website
8. Run docker pull + the image name pushed on another machine to run the modified image on other machines
Mount data to external modification
In order not to enter the virtual machine running the image when you want to modify the image, mount the file directory outside the virtual machine to the directory of the virtual machine to realize the external modification of data
Add the - v parameter to the regular docker startup command to mount
docker run --name=mynginx -d \ --restart=always -p 80:88 -v /data/html:/usr/share/nginx/html:ro nginx
*The mounting of this file is equivalent to directly replacing the file path inside the virtual machine. When using it, pay attention to add the files required for the image in the mounted path and modify it
*-v you can add multiple - v parameters to mount multiple files
After mounting, you can modify the image on the host
Example: deploying redis
Start redis with the redis.conf parameter configuration file edited by yourself, persist the data to the specified directory, and mount the redis parameter configuration file and the persistence directory respectively
In the redis.conf configuration file of redis, simply write appendonly yes to start persistence for testing
docker run -v /root/test/redis/redis.conf:/etc/redis/redis.conf \ -v /root/test/redis/data:/data -d --name=myredis \ -p 6379:6379 redis redis-server /etc/redis/redis.conf