$ docker
$ docker version
$ docker info
$ docker container run -it -p 80:80 nginx
$ docker container run -d -p 80:80 nginx
$ docker container run -d -p 80:80 nginx
$ docker container run -d -p 80:80 --name nginx-server nginx
- Looked for image called nginx in image cache
- If not found in cache, it looks to the default image repo on Dockerhub
- Pulled it down (latest version), stored in the image cache
- Started it in a new container
- We specified to take port 80- on the host and forward to port 80 on the container
- We could do "$ docker container run --publish 8000:80 --detach nginx" to use port 8000
- We can specify versions like "nginx:1.09"
$ docker container ls
OR
$ docker ps
$ docker container ls -a
$ docker container stop [ID]
$ docker stop $(docker ps -aq)
$ docker container rm [ID]
$ docker container rm -f [ID]
$ docker container rm [ID] [ID] [ID]
$ docker rm $(docker ps -aq)
$ docker container logs [NAME]
$ docker container top [NAME]
Docker containers are often compared to virtual machines but they are actually just processes running on your host os. In Windows/Mac, Docker runs in a mini-VM so to see the processes youll need to connect directly to that. On Linux however you can run "ps aux" and see the processes directly
$ docker image ls
$ docker pull [IMAGE]
$ docker image rm [IMAGE]
$ docker rmi $(docker images -a -q)
- Images are app bianaries and dependencies with meta data about the image data and how to run the image
- Images are no a complete OS. No kernel, kernel modules (drivers)
- Host provides the kernel, big difference between VM
NGINX:
$ docker container run -d -p 80:80 --name nginx nginx (-p 80:80 is optional as it runs on 80 by default)
APACHE:
$ docker container run -d -p 8080:80 --name apache httpd
MONGODB:
$ docker container run -d -p 27017:27017 --name mongo mongo
MYSQL:
$ docker container run -d -p 3306:3306 --name mysql --env MYSQL_ROOT_PASSWORD=123456 mysql
$ docker container inspect [NAME]
$ docker container inspect --format '{{ .NetworkSettings.IPAddress }}' [NAME]
$ docker container stats [NAME]
$ docker container run -it --name [NAME] nginx bash
- i = interactive Keep STDIN open if not attached
- t = tty - Open prompt
For Git Bash, use "winpty"
$ winpty docker container run -it --name [NAME] nginx bash
$ docker container run -it --name ubuntu ubuntu
(no bash because ubuntu uses bash by default)
$ docker container run --rm -it --name [NAME] ubuntu
$ docker container start -ai ubuntu
$ docker container exec -it mysql bash
$ docker container run -it alpine sh
(use sh because it does not include bash) (alpine uses apk for its package manager - can install bash if you want)
In order to limit the amount of memory a docker container process can use, simply set the -m [memory amount] flag with the limit.
To run a container with memory limited to 256 MBs:
Example: docker run -name [name] -m [Memory (int)][memory unit (b, k, m or g)] -d (to run not to attach) -p (to set access and expose ports) [image ID]
$ docker run -m 64m -d -p 8082:80 tutum/wordpress
$ docker container port [NAME]
$ docker network ls
$ docker network inspect [NETWORK_NAME]
("bridge" is default)
$ docker network create [NETWORK_NAME]
or
$ docker network create --driver bridge [NETWORK_NAME]
$ docker run -d --net=[NETWORK_NAME] --name mongodb mongo
$ docker container run -d --name [NAME] --network [NETWORK_NAME] nginx
$ docker network connect [NETWORK_NAME] [CONTAINER_NAME]
$ docker network disconnect [NETWORK_NAME] [CONTAINER_NAME]
$ docker network disconnect
To remove the network by name or id, multiple can be deleted:
$ docker network rm [NETWORK_NAME] [NETWORK_NAME]
$ docker image ls
Youll see that each image has a tag
$ docker image tag nginx btraversy/nginx
$ docker image push bradtraversy/nginx
$ docker login
$ docker image tag bradtraversy/nginx bradtraversy/nginx:testing
- FROM - The os used. Common is alpine, debian, ubuntu
- ENV - Environment variables
- RUN - Run commands/shell scripts, etc
- EXPOSE - Ports to expose
- CMD - Final command run when you launch a new container from image
- WORKDIR - Sets working directory (also could use 'RUN cd /some/path')
- COPY # Copies files from host to container
$ docker image build -t [REPONAME] .
$ DOCKER_BUILDKIT=1 docker image build -t [REPONAME] .
- If you re-run the build, it will be quick because everythging is cached.
- If you change one line and re-run, that line and everything after will not be cached
- Keep things that change the most toward the bottom of the Dockerfile
FROM nginx:latest # Extends nginx so everything included in that image is included here
WORKDIR /usr/share/nginx/html
COPY index.html index.html
$ docker image build -t nginx-website
$ docker container run -p 80:80 --rm nginx-website
$ docker image tag nginx-website:latest btraversy/nginx-website:latest
$ docker image push bradtraversy/nginx-website
$ docker volume ls
$ docker volume prune
$ docker pull mysql
$ docker image inspect mysql
$ docker container run -d --name mysql -e MYSQL_ALLOW_EMPTY_PASSWORD=True mysql
$ docker container inspect mysql
- You will also see the volume under mounts
- Container gets its own uniqe location on the host to store that data
- Source: xxx is where it lives on the host
$ docker volume ls
There is no way to tell volumes apart for instance with 2 mysql containers, so we used named volumes
$ docker container run -d --name mysql -e MYSQL_ALLOW_EMPTY_PASSWORD=True -v mysql-db:/var/lib/mysql mysql
docker volume inspect mysql-db
- Can not use in Dockerfile, specified at run time (uses -v as well)
- ... run -v /Users/brad/stuff:/path/container (mac/linux)
- ... run -v //c/Users/brad/stuff:/path/container (windows)
TIP: Instead of typing out local path, for working directory use $(pwd):/path/container - On windows may not work unless you are in your users folder
$ docker container run -p 80:80 -v $(pwd):/usr/share/nginx/html nginx
$ docker container exec -it nginx bash
$ cd /usr/share/nginx/html
$ ls -al
$ touch test.txt
$ docker run -d --name my-postgres postgres
now Link it with dot net
$ docker run -d -p 5000:5000 --link my-postgres:postgres btree/dotnet
- Configure relationships between containers
- Save our docker container run settings in easy to read file
- 2 Parts: YAML File (docker.compose.yml) + CLI tool (docker-compose)
- containers
- networks
- volumes
version: '2'
# same as
# docker run -p 80:4000 -v $(pwd):/site bretfisher/jekyll-serve
services:
jekyll:
image: bretfisher/jekyll-serve
volumes:
- .:/site
ports:
- '80:4000'
version: '3'
services:
app:
container_name: docker-node-mongo
restart: always
build: .
ports:
- '80:3000'
links:
- mongo
mongo:
container_name: mongo
image: mongo
ports:
- '27017:27017'
docker-compose up
docker-compose up -d
docker-compose down