目录
- Docker基础入门
- Docker基本命令
- Dockerfile用法
- 制作小镜像上
- 多阶段制作小镜像下
- Scratch空镜像
Docker基础入门
Docker:它是一个开源的软件项目,在Linux操作系统上,docker提供了一个额外的软件抽象层及操作系统层虚拟化的自动管理机制。
物理机:
- 安装系统
- 依赖环境
- 应用程序
- 多个物理机提高并发量
虚拟机:
- 把一个物理机虚拟机虚拟成多个机器
- 把依赖环境打成一个系统的模板
容器化:
- 镜像基础(依赖环境的镜像;根据基础镜像放入自己的代码或者包;按层存储)
- 启动时间特别快,秒级启动
容器:把自己的应用程序,根据某个依赖的基础镜像,生成一个应用程序镜像
应用程序镜像,可以运行在任何部署了Docker环境的机器上。
Docker基本命令
- 查看Docker版本
- Docker详细信息
- 搜索镜像
- 拉取镜像
- 推送镜像
- 启动镜像
- 查看容器
- 查看日志
- 进入容器
- 复制文件
- 删除镜像
- 修改记录
- 保存状态
查看Docker版本
[root@k8s-master01 ~]# docker version
Client: Docker Engine - Community
Version: 20.10.7
API version: 1.40
Go version: go1.13.15
Git commit: f0df350
Built: Wed Jun 2 11:58:10 2021
OS/Arch: linux/amd64
Context: default
Experimental: true
Server: Docker Engine - Community
Engine:
Version: 19.03.15
API version: 1.40 (minimum version 1.12)
Go version: go1.13.15
Git commit: 99e3ed8919
Built: Sat Jan 30 03:16:33 2021
OS/Arch: linux/amd64
Experimental: false
containerd:
Version: 1.4.6
GitCommit: d71fcd7d8303cbf684402823e425e9dd2e99285d
runc:
Version: 1.0.0-rc95
GitCommit: b9ee9c6314599f1b4a7f497e1f1f856fe433d3b7
docker-init:
Version: 0.18.0
GitCommit: fec3683
Docker详细信息
[root@k8s-master01 ~]# docker info
Client:
Context: default
Debug Mode: false # 可以按需修改
Plugins:
app: Docker App (Docker Inc., v0.9.1-beta3)
buildx: Build with BuildKit (Docker Inc., v0.5.1-docker)
scan: Docker Scan (Docker Inc., v0.8.0)
Server:
Containers: 6
Running: 4
Paused: 0
Stopped: 2
Images: 9
Server Version: 19.03.15
Storage Driver: overlay2 # 官方推荐的存储驱动,要求文件系统是xfs,必须支持d_type(目录条目类型,内核上的一个数据,安装系统的时候必须把ftype设置为1,不然的话很影响docker的性能),目前流行的Storage Driver有aufs(ubuntu支持)、overlay、brtfs
Backing Filesystem: xfs
Supports d_type: true
Native Overlay Diff: true
Logging Driver: json-file # docker日志的存储类型,json-file会存储在本地,目录在Docker Root Dir
Cgroup Driver: systemd
Plugins:
Volume: local
Network: bridge host ipvlan macvlan null overlay
Log: awslogs fluentd gcplogs gelf journald json-file local logentries splunk syslog
Swarm: inactive # 容器编排工具,inactive未启动
Runtimes: runc
Default Runtime: runc # docker运行的核心
Init Binary: docker-init
containerd version: d71fcd7d8303cbf684402823e425e9dd2e99285d
runc version: b9ee9c6314599f1b4a7f497e1f1f856fe433d3b7
init version: fec3683
Security Options:
seccomp
Profile: default
Kernel Version: 4.19.12-1.el7.elrepo.x86_64 # 内核版本低于3.18不能使用overlay2存储驱动,linux3.18才加入的
Operating System: CentOS Linux 7 (Core)
OSType: linux
Architecture: x86_64
CPUs: 2
Total Memory: 1.923GiB
Name: k8s-master01
ID: PYSL:2OAV:4C7N:WAI3:7G3J:IBR5:6BXI:7OEG:YNCL:6RAR:3CTF:CTDD
Docker Root Dir: /var/lib/docker # 这个目录可以修改,生产环境推荐使用ssd硬盘存储,挂载在该目录,可以提高docker的性能,条件不足的情况下,最好使用一个单独的磁盘进行挂载,不要和宿主机用同一个磁盘
Debug Mode: false
Registry: https://index.docker.io/v1/ # 官方镜像仓库,访问较慢,可以修改为自己的镜像仓库
Labels:
Experimental: false
Insecure Registries: # 如果使用的非官方镜像仓库地址不是https,需要把链接加入到此非安全列表
127.0.0.0/8
Live Restore Enabled: false # 生产环境中需要打开,重启docker进程不会重启正在运行的容器,如果容器没有设置自动重启就会被关闭,不会重启
查看d_type信息(ftype=1说明系统支持d_type)
[root@k8s-master01 ~]# xfs_info /
meta-data=/dev/mapper/centos-root isize=512 agcount=4, agsize=1113856 blks
= sectsz=512 attr=2, projid32bit=1
= crc=1 finobt=0 spinodes=0
data = bsize=4096 blocks=4455424, imaxpct=25
= sunit=0 swidth=0 blks
naming =version 2 bsize=4096 ascii-ci=0 ftype=1
log =internal bsize=4096 blocks=2560, version=2
= sectsz=512 sunit=0 blks, lazy-count=1
realtime =none extsz=4096 blocks=0, rtextents=0
进入Docker Root Dir,进入任何一个容器目录既可查看对应容器的日志,每次docker重启的时候日志就会被清除
[root@k8s-master01 ~]# cd /var/lib/docker/containers/
[root@k8s-master01 containers]# ll
total 0
drwx-----x 4 root root 165 Jul 12 10:54 10b58a593a5f417d466fbb2eba54c6ac0e8322a3712cbc6eb46f9cae5b48e4d8
[root@k8s-master01 10b58a593a5f417d466fbb2eba54c6ac0e8322a3712cbc6eb46f9cae5b48e4d8]# ll
total 16
-rw-r----- 1 root root 4042 Jul 12 10:53 10b58a593a5f417d466fbb2eba54c6ac0e8322a3712cbc6eb46f9cae5b48e4d8-json.log
Insecure Registries目录
[root@k8s-master01 ~]# vim /etc/docker/daemon.json
{
"registry-mirrors": [
"https://registry.docker-cn.com",
"http://hub-mirror.c.163.com",
"https://docker.mirrors.ustc.edu.cn"
],
"exec-opts": ["native.cgroupdriver=systemd"],
"max-concurrent-downloads": 10, # 并发下载的线程数
"max-concurrent-uploads": 5, # 并发上传的线程数
"log-opts": {
"max-size": "300m", # 限制日志文件最大容量,超过则分割
"max-file": "2" # 日志保存最大数量
},
"live-restore": true # 更改docker配置之后需要重启docker才能生效,这个参数可以使得重启docker不影响正在运行的容器进程
}
搜索镜像
[root@k8s-master01 ~]# docker search centos
NAME DESCRIPTION STARS OFFICIAL AUTOMATED
centos The official build of CentOS. 6631 [OK]
ansible/centos7-ansible Ansible on Centos7 134 [OK]
consol/centos-xfce-vnc Centos container with "headless" VNC session… 129 [OK]
jdeathe/centos-ssh OpenSSH / Supervisor / EPEL/IUS/SCL Repos - … 118 [OK]
centos/systemd systemd enabled base container. 100 [OK]
centos/mysql-57-centos7 MySQL 5.7 SQL database server 88
imagine10255/centos6-lnmp-php56 centos6-lnmp-php56 58 [OK]
tutum/centos Simple CentOS docker image with SSH access 48
centos/postgresql-96-centos7 PostgreSQL is an advanced Object-Relational … 45
jdeathe/centos-ssh-apache-php Apache PHP - CentOS. 31 [OK]
kinogmt/centos-ssh CentOS with SSH 29 [OK]
guyton/centos6 From official centos6 container with full up… 10 [OK]
nathonfowlie/centos-jre Latest CentOS image with the JRE pre-install… 8 [OK]
centos/tools Docker image that has systems administration… 7 [OK]
drecom/centos-ruby centos ruby 6 [OK]
mamohr/centos-java Oracle Java 8 Docker image based on Centos 7 3 [OK]
darksheer/centos Base Centos Image -- Updated hourly 3 [OK]
miko2u/centos6 CentOS6 日本語環境 2 [OK]
amd64/centos The official build of CentOS. 2
dokken/centos-7 CentOS 7 image for kitchen-dokken 2
indigo/centos-maven Vanilla CentOS 7 with Oracle Java Developmen… 2 [OK]
mcnaughton/centos-base centos base image 1 [OK]
blacklabelops/centos CentOS Base Image! Built and Updates Daily! 1 [OK]
starlabio/centos-native-build Our CentOS image for native builds 0 [OK]
smartentry/centos centos with smartentry 0 [OK]
OFFICIAL的值是ok代表官方容器,一般都是使用官方的
拉取镜像
alpine可以作为基础镜像,拉取镜像的时候如果本地已经存在,则不会重复拉取
可以通过官网搜索镜像:https://hub.docker.com/
拉取镜像,如果本地已有该镜像,则不会重复拉取
[root@k8s-master01 ~]# docker pull alpine:latest
latest: Pulling from library/alpine
5843afab3874: Pull complete
Digest: sha256:234cb88d3020898631af0ccbbcca9a66ae7306ecd30c9720690858c1b007d2a0
Status: Downloaded newer image for alpine:latest
docker.io/library/alpine:latest
默认拉取官方镜像,如果需要拉取指定镜像,需要添加地址,版本号
推送镜像
查看镜像
[root@k8s-master01 ~]# docker images
REPOSITORY TAG IMAGE ID CREATED SIZE
nginx latest 4cdc5dd7eaad 7 days ago 133MB
kubernetesui/dashboard v2.3.1 e1482a24335a 3 weeks ago 220MB
alpine latest d4ff818577bc 4 weeks ago 5.6MB
registry.cn-beijing.aliyuncs.com/dotbalo/node v3.15.3 d45bf977dfbf 10 months ago 262MB
registry.cn-beijing.aliyuncs.com/dotbalo/pod2daemon-flexvol v3.15.3 963564fb95ed 10 months ago 22.8MB
registry.cn-beijing.aliyuncs.com/dotbalo/cni v3.15.3 ca5564c06ea0 10 months ago 110MB
kubernetesui/dashboard v2.0.3 503bc4b7440b 12 months ago 225MB
registry.cn-beijing.aliyuncs.com/dotbalo/coredns 1.7.0 bfe3a36ebd25 13 months ago 45.2MB
registry.cn-beijing.aliyuncs.com/dotbalo/metrics-scraper v1.0.4 86262685d9ab 15 months ago 36.9MB
registry.cn-hangzhou.aliyuncs.com/google_containers/pause-amd64 3.2 80d28bedfe5d 17 months ago 683kB
镜像打标签到自己的镜像仓库
[root@k8s-master01 ~]# docker tag registry.cn-beijing.aliyuncs.com/dotbalo/pod2daemon-flexvol:v3.15.3 mingsonzheng/pod2daemon-flexvol:v3.15.3
登录远程镜像仓库
[root@k8s-master01 ~]# docker login
Login with your Docker ID to push and pull images from Docker Hub. If you don't have a Docker ID, head over to https://hub.docker.com to create one.
Username: mingsonzheng
Password:
WARNING! Your password will be stored unencrypted in /root/.docker/config.json.
Configure a credential helper to remove this warning. See
https://docs.docker.com/engine/reference/commandline/login/#credentials-store
Login Succeeded
登录到指定仓库
docker login xxx.com
推送镜像
[root@k8s-master01 ~]# docker push mingsonzheng/pod2daemon-flexvol:v3.15.3
The push refers to repository [docker.io/mingsonzheng/pod2daemon-flexvol]
f0e55d2e215d: Pushed
7b2f85666007: Pushed
752045c6df15: Pushed
ca07dc9dd06e: Pushed
1d0352c1c217: Pushed
540c65dd0455: Pushed
48855504bcc3: Pushed
v3.15.3: digest: sha256:6bd1246d0ea1e573a6a050902995b1666ec0852339e5bda3051f583540361b55 size: 1788
启动镜像
- 前台启动
- 后台启动
前台启动,如果本地没有镜像会先拉取
[root@k8s-master01 ~]# docker run -ti centos:8 bash
Unable to find image 'centos:8' locally
8: Pulling from library/centos
7a0437f04f83: Pull complete
Digest: sha256:5528e8b1b1719d34604c87e11dcd1c0a20bedf46e83b5632cdeac91b8c04efc1
Status: Downloaded newer image for centos:8
[root@a4cb8f5d6bd5 /]#
后台启动
[root@k8s-master01 ~]# docker run -d centos:8 bash
617a1213ae5ce5c4bed0716fd5b3212b25ed7bcc2099ba329db54cccd3c0b8d5
[root@k8s-master01 ~]#
查看容器
查看正在运行的容器
[root@k8s-master01 ~]# docker ps
CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES
617a1213ae5c centos:8 "bash" About a minute ago Exited (0) About a minute ago vigilant_boyd
04b21dff00f5 centos:8 "bash" 2 minutes ago Exited (0) 2 minutes ago objective_satoshi
a4cb8f5d6bd5 centos:8 "bash" 5 minutes ago Exited (127) 2 minutes ago reverent_elgamal
查看所有状态的容器
[root@k8s-master01 ~]# docker ps -a
CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES
1d906297395d centos:8 "bash" 2 minutes ago Exited (0) 2 minutes ago wonderful_hermann
617a1213ae5c centos:8 "bash" 22 minutes ago Exited (0) 22 minutes ago vigilant_boyd
04b21dff00f5 centos:8 "bash" 22 minutes ago Exited (0) 22 minutes ago objective_satoshi
a4cb8f5d6bd5 centos:8 "bash" 26 minutes ago Exited (127) 22 minutes ago reverent_elgamal
a3db403fd0be kubernetesui/dashboard "/dashboard --insecu…" 46 hours ago Up 46 hours k8s_kubernetes-dashboard_kubernetes-dashboard-67484c44f6-brz2z_kubernetes-dashboard_bb016e8e-a4b1-48e6-912c-022addfb72bf_0
8cd76e9123b3 registry.cn-hangzhou.aliyuncs.com/google_containers/pause-amd64:3.2 "/pause" 46 hours ago Up 46 hours k8s_POD_kubernetes-dashboard-67484c44f6-brz2z_kubernetes-dashboard_bb016e8e-a4b1-48e6-912c-022addfb72bf_0
dbec965d88da registry.cn-beijing.aliyuncs.com/dotbalo/node "start_runit" 2 days ago Up 2 days k8s_calico-node_calico-node-gn8ws_kube-system_16eeb29c-3383-4e62-b84c-40bf3dae7d07_0
e9869813fc1b registry.cn-beijing.aliyuncs.com/dotbalo/pod2daemon-flexvol "/usr/local/bin/flex…" 2 days ago Exited (0) 2 days ago k8s_flexvol-driver_calico-node-gn8ws_kube-system_16eeb29c-3383-4e62-b84c-40bf3dae7d07_0
10b58a593a5f registry.cn-beijing.aliyuncs.com/dotbalo/cni "/install-cni.sh" 2 days ago Exited (0) 2 days ago k8s_install-cni_calico-node-gn8ws_kube-system_16eeb29c-3383-4e62-b84c-40bf3dae7d07_0
610909969be1 registry.cn-hangzhou.aliyuncs.com/google_containers/pause-amd64:3.2 "/pause" 2 days ago Up 2 days k8s_POD_calico-node-gn8ws_kube-system_16eeb29c-3383-4e62-b84c-40bf3dae7d07_0
查看正在运行的容器的id
[root@k8s-master01 ~]# docker ps -q
a3db403fd0be
8cd76e9123b3
dbec965d88da
610909969be1
查看所有的容器的id
[root@k8s-master01 ~]# docker ps -aq
1d906297395d
617a1213ae5c
04b21dff00f5
a4cb8f5d6bd5
a3db403fd0be
8cd76e9123b3
dbec965d88da
e9869813fc1b
10b58a593a5f
610909969be1
查看日志
[root@k8s-master01 ~]# docker logs -f a3db403fd0be
对应日志路径
/var/lib/docker/containers/a3db403fd0be5eacb8aa1769cc8d215da93eb74da84262aa1aac11551c37a84d
进入容器
启动容器
[root@k8s-master01 ~]# docker run -ti nginx:1.14.2 sh
#
Xshell 7新建一个链接到服务器
查看运行中的容器
[root@k8s-master01 ~]# docker ps
CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES
2ef850f11136 nginx:1.14.2 "sh" 29 seconds ago Up 28 seconds 80/tcp funny_leavitt
a3db403fd0be kubernetesui/dashboard "/dashboard --insecu…" 2 days ago Up 2 days k8s_kubernetes-dashboard_kubernetes-dashboard-67484c44f6-brz2z_kubernetes-dashboard_bb016e8e-a4b1-48e6-912c-022addfb72bf_0
8cd76e9123b3 registry.cn-hangzhou.aliyuncs.com/google_containers/pause-amd64:3.2 "/pause" 2 days ago Up 2 days k8s_POD_kubernetes-dashboard-67484c44f6-brz2z_kubernetes-dashboard_bb016e8e-a4b1-48e6-912c-022addfb72bf_0
dbec965d88da registry.cn-beijing.aliyuncs.com/dotbalo/node "start_runit" 3 days ago Up 3 days k8s_calico-node_calico-node-gn8ws_kube-system_16eeb29c-3383-4e62-b84c-40bf3dae7d07_0
610909969be1 registry.cn-hangzhou.aliyuncs.com/google_containers/pause-amd64:3.2 "/pause" 3 days ago Up 3 days k8s_POD_calico-node-gn8ws_kube-system_16eeb29c-3383-4e62-b84c-40bf3dae7d07_0
进入容器
[root@k8s-master01 ~]# docker exec -ti 2ef850f11136 sh
# ls
bin boot dev etc home lib lib64 media mnt opt proc root run sbin srv sys tmp usr var
# exit
[root@k8s-master01 ~]#
复制文件
后台启动nginx
[root@k8s-master01 ~]# docker run -tid -p 12345:80 nginx:latest
b75ad319e61326e05905d41d67616112be51af4d1234f51bef5d21807e9bc42d
[root@k8s-master01 ~]# docker ps
CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES
b75ad319e613 nginx:latest "/docker-entrypoint.…" 8 seconds ago Up 6 seconds 0.0.0.0:12345->80/tcp upbeat_brahmagupta
访问地址:http://192.168.232.128:12345/
进入容器
[root@k8s-master01 ~]# docker exec -ti b75ad319e613 bash
root@b75ad319e613:/# cd /usr/share/nginx/html/
root@b75ad319e613:/usr/share/nginx/html# ls
50x.html index.html
root@b75ad319e613:/usr/share/nginx/html# exit
exit
复制文件到容器中
[root@k8s-master01 ~]# echo "test cp" > index.html
[root@k8s-master01 ~]# docker cp index.html b75ad319e613:/usr/share/nginx/html/
访问地址:http://192.168.232.128:12345/
复制容器中的文件到本地
[root@k8s-master01 ~]# docker cp b75ad319e613:/usr/share/nginx/html/index.html .
删除镜像
[root@k8s-master01 ~]# docker images
REPOSITORY TAG IMAGE ID CREATED SIZE
nginx latest 4cdc5dd7eaad 8 days ago 133MB
[root@k8s-master01 ~]# docker rmi 4cdc5dd7eaad
Error response from daemon: conflict: unable to delete 4cdc5dd7eaad (cannot be forced) - image is being used by running container b75ad319e613
# 删除容器
[root@k8s-master01 ~]# docker rm b75ad319e613
Error response from daemon: You cannot remove a running container b75ad319e61326e05905d41d67616112be51af4d1234f51bef5d21807e9bc42d. Stop the container before attempting removal or force remove
# 停止容器运行
[root@k8s-master01 ~]# docker stop b75ad319e613
b75ad319e613
[root@k8s-master01 ~]# docker rm b75ad319e613
b75ad319e613
[root@k8s-master01 ~]# docker rmi 4cdc5dd7eaad
Error response from daemon: conflict: unable to delete 4cdc5dd7eaad (must be forced) - image is being used by stopped container cd2a20f47459
[root@k8s-master01 ~]# docker rm cd2a20f47459
cd2a20f47459
[root@k8s-master01 ~]# docker rmi 4cdc5dd7eaad
Untagged: nginx:latest
Untagged: nginx@sha256:353c20f74d9b6aee359f30e8e4f69c3d7eaea2f610681c4a95849a2fd7c497f9
Deleted: sha256:4cdc5dd7eaadff5080649e8d0014f2f8d36d4ddf2eff2fdf577dd13da85c5d2f
Deleted: sha256:63d268dd303e176ba45c810247966ff8d1cb9a5bce4a404584087ec01c63de15
Deleted: sha256:b27eb5bbca70862681631b492735bac31d3c1c558c774aca9c0e36f1b50ba915
Deleted: sha256:435c6dad68b58885ad437e5f35f53e071213134eb9e4932b445eac7b39170700
Deleted: sha256:bdf28aff423adfe7c6cb938eced2f19a32efa9fa3922a3c5ddce584b139dc864
Deleted: sha256:2c78bcd3187437a7a5d9d8dbf555b3574ba7d143c1852860f9df0a46d5df056a
Deleted: sha256:764055ebc9a7a290b64d17cf9ea550f1099c202d83795aa967428ebdf335c9f7
修改记录
[root@k8s-master01 ~]# docker images
registry.cn-hangzhou.aliyuncs.com/google_containers/pause-amd64 3.2 80d28bedfe5d 17 months ago 683kB
[root@k8s-master01 ~]# docker history registry.cn-hangzhou.aliyuncs.com/google_containers/pause-amd64:3.2
IMAGE CREATED CREATED BY SIZE COMMENT
80d28bedfe5d 17 months ago ENTRYPOINT ["/pause"] 0B buildkit.dockerfile.v0
<missing> 17 months ago ADD bin/pause-amd64 /pause # buildkit 683kB buildkit.dockerfile.v0
<missing> 17 months ago ARG ARCH 0B buildkit.dockerfile.v0
保存状态
假设我们复制了文件到容器中,想要保存容器状态需要使用 docker commit
# 拉取镜像
[root@k8s-master01 ~]# docker pull nginx:latest
latest: Pulling from library/nginx
b4d181a07f80: Pull complete
66b1c490df3f: Pull complete
d0f91ae9b44c: Pull complete
baf987068537: Pull complete
6bbc76cbebeb: Pull complete
32b766478bc2: Pull complete
Digest: sha256:353c20f74d9b6aee359f30e8e4f69c3d7eaea2f610681c4a95849a2fd7c497f9
Status: Downloaded newer image for nginx:latest
docker.io/library/nginx:latest
# 后台启动容器
[root@k8s-master01 ~]# docker run -tid -p 12345:80 nginx:latest
8ddcc9d728450c9398579155e793aca873713632d72a8402e5684ebb6f613437
# 获取容器ID
[root@k8s-master01 ~]# docker ps
CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES
8ddcc9d72845 nginx:latest "/docker-entrypoint.…" 28 seconds ago Up 27 seconds 0.0.0.0:12345->80/tcp confident_leavitt
# 复制文件到容器中
[root@k8s-master01 ~]# docker cp index.html 8ddcc9d72845:/usr/share/nginx/html/
# 提交修改记录
[root@k8s-master01 ~]# docker commit -a "mingsonzheng" -m "update index.html" 8ddcc9d72845 nginx:commit
sha256:f62c69e4853e45951b1b83eec2432e48e436d65a14a62b087e5fe6c7c9398771
# 查看镜像
[root@k8s-master01 ~]# docker images
REPOSITORY TAG IMAGE ID CREATED SIZE
nginx commit f62c69e4853e 40 seconds ago 133MB
# 启动容器,加上--rm,则关闭之后会删除,不会产生大量状态为Exited的容器
[root@k8s-master01 ~]# docker run -ti --rm nginx:commit bash
root@c657e57078c1:/# cd /usr/share/nginx/html/
root@c657e57078c1:/usr/share/nginx/html# ls
50x.html index.html
root@c657e57078c1:/usr/share/nginx/html# exit
exit
# 查看容器,该容器不会出现在列表中
[root@k8s-master01 ~]# docker ps
Dockerfile用法
- Dockerfile指令
- Dockerfile用法
Dockerfile指令
- FROM:继承基础镜像
- MAINTAINER:镜像制作作者信息
- RUN:用来执行shell命令
- EXPOSE:暴露端口号
- CMD:启动容器默认执行的命令
- ENTRYPOINT:启动容器真正执行的命令
- VOLUME:创建挂载点
- ENV:配置环境变量
- ADD:添加文件到容器
- COPY:复制文件到容器
- WORKDIR:设置容器的工作目录
- USER:容器使用的用户
nginx官方Dockerfile:https://github.com/nginxinc/docker-nginx/blob/master/Dockerfile-alpine.template
继承alpine基础镜像
FROM alpine:%%ALPINE_VERSION%%
MAINTAINER:镜像制作作者信息,新版本推荐使用LABEL
LABEL maintainer="NGINX Docker Maintainers <docker-maint@nginx.com>"
执行shell命令
RUN set -x
apk是alpine安装包的工具
apk add --no-cache --virtual .cert-deps
暴露端口号
EXPOSE 80
启动容器默认执行的命令
CMD ["nginx", "-g", "daemon off;"]
CMD命令可以被覆盖,比如bash
[root@k8s-master01 ~]# docker run -ti --rm nginx:commit bash
CMD 和 ENTRYPOINT 必须有一个,如果两者都有的话,ENTRYPOINT 就是启动命令,而 CMD 就是它的参数
ENTRYPOINT 对应 k8s 的 COMMAND
CMD 对应 k8s 的 arg
VOLUME 是一个匿名的挂载点,Docker 不建议保留容器产生的文件,但是有一些文件是必须保留的,比如数据库,我们可以为其创建一个挂载点,将宿主机的目录挂载到容器里面,这样就可以保留它的数据
配置环境变量
ENV NGINX_VERSION %%NGINX_VERSION%%
ENV NJS_VERSION %%NJS_VERSION%%
ENV PKG_RELEASE %%PKG_RELEASE%%
但是不建议写在 Dockerfile 中,建议通过命令行 -e 传入参数,独立到镜像之外,可以动态修改
[root@k8s-master01 ~]# docker run -ti -e a=b -e c=d
ADD:复制文件到容器,如果复制压缩文件会自动解压
COPY:复制文件到容器,不会解压,直接复制
WORKDIR:设置容器的工作目录,如果有设置则进入容器时会进入到该目录下
USER:容器使用的用户,系统默认用户是root,但是在业务容器容器化的时候不建议使用root,因为不安全,可能会造成对宿主机有危险的操作
[root@k8s-master01 ~]# whoami
root
Dockerfile用法
新增Dockerfile
[root@k8s-master01 ~]# mkdir dockerfiles
[root@k8s-master01 ~]# cd dockerfiles/
[root@k8s-master01 dockerfiles]# vim Dockerfile
# 添加以下内容
FROM centos:8
LABEL maintainer="test dockerfile"
LABEL test=dockerfile
RUN useradd mingsonzheng
RUN mkdir /opt/mingsonzheng
在当前目录下构建镜像
[root@k8s-master01 dockerfiles]# docker build -t centos:user .
Sending build context to Docker daemon 2.048kB
Step 1/5 : FROM centos:8
---> 300e315adb2f
Step 2/5 : LABEL maintainer="test dockerfile"
---> Running in cccd20f9b0f5
Removing intermediate container cccd20f9b0f5
---> f6dc016add09
Step 3/5 : LABEL test=dockerfile
---> Running in e4d1962c630b
Removing intermediate container e4d1962c630b
---> 3977dc88a2f2
Step 4/5 : RUN useradd mingsonzheng
---> Running in d3905f2c7dc4
Removing intermediate container d3905f2c7dc4
---> 30968f5c2285
Step 5/5 : RUN mkdir /opt/mingsonzheng
---> Running in 811c4ceb3e70
Removing intermediate container 811c4ceb3e70
---> 0cd811c2fadf
Successfully built 0cd811c2fadf
Successfully tagged centos:user
启动镜像
[root@k8s-master01 dockerfiles]# docker run -ti --rm centos:user bash
[root@828f29ae30bf /]#
调试的时候一定要加上 --rm,不然会有很多Exited记录
查看用户
[root@828f29ae30bf /]# cat /etc/passwd
mingsonzheng:x:1000:1000::/home/mingsonzheng:/bin/bash
制作完镜像之后,如果调试没有问题,记得推送到镜像仓库
如果测试通过之后,需要优化一下RUN语句,没有必要分开写,因为docker有缓存机制,假设有20个RUN语句,执行第18个的时候失败了,前面17个执行成功的没有必要再执行一次,修改之后直接从第18个RUN语句继续执行
[root@k8s-master01 dockerfiles]# vim Dockerfile
# 修改为以下内容
FROM centos:8
LABEL maintainer="test dockerfile"
LABEL test=dockerfile
#RUN useradd mingsonzheng
#RUN mkdir /opt/mingsonzheng
RUN useradd mingsonzheng && /opt/mingsonzheng
这样docker的层级会少一层,存储大小会小一点
CMD:启动容器默认执行的命令
[root@k8s-master01 dockerfiles]# vim Dockerfile
# 修改为以下内容
FROM centos:8
LABEL maintainer="test dockerfile"
LABEL test=dockerfile
RUN useradd mingsonzheng
RUN mkdir /opt/mingsonzheng
CMD [ "sh", "-c", "echo 1"]
#RUN useradd mingsonzheng && /opt/mingsonzheng
构建镜像
[root@k8s-master01 dockerfiles]# docker build -t centos:cmd .
如果文件不在当前目录,可以使用-f
[root@k8s-master01 dockerfiles]# mkdir t
[root@k8s-master01 dockerfiles]# cp Dockerfile t
[root@k8s-master01 dockerfiles]# docker build -t centos:cmd -f t/Dockerfile
查看镜像
[root@k8s-master01 dockerfiles]# docker images
REPOSITORY TAG IMAGE ID CREATED SIZE
centos cmd a008df4670f8 2 minutes ago 210MB
启动容器
[root@k8s-master01 dockerfiles]# docker run -ti --rm centos:cmd
1
可以看到打印了一个1就退出了
使用bash覆盖CMD命令
[root@k8s-master01 dockerfiles]# docker run -ti --rm centos:cmd bash
[root@924006c7dc0e /]# exit
exit
可以看到它就不打印了,覆盖了CMD命令
在k8s中不建议将CMD命令打到镜像中,因为我们遵循的构建策略是一次构建到处运行,启动的命令可能不是统一的,可以指定配置文件让启动命令根据配置文件变成不同环境的命令,这样可以使用k8s的arg或者cmd去覆盖掉它的启动参数,所以CMD可以不打到镜像中,当然也可以使用变量的注入方式
使用ENTRYPIOINT,CMD可以被覆盖,如果有ENTRYPIOINT的话,CMD就是ENTRYPIOINT的参数
[root@k8s-master01 dockerfiles]# vim Dockerfile
# 修改为以下内容
FROM centos:8
LABEL maintainer="test dockerfile"
LABEL test=dockerfile
RUN useradd mingsonzheng
RUN mkdir /opt/mingsonzheng
ENTRYPOINT ["echo"]
CMD [ "3"]
#RUN useradd mingsonzheng && /opt/mingsonzheng
打包镜像
[root@k8s-master01 dockerfiles]# docker build -t centos:ep .
运行容器
[root@k8s-master01 dockerfiles]# docker run -ti --rm centos:ep 10
10
[root@k8s-master01 dockerfiles]# docker run -ti --rm centos:ep mingsonzheng
mingsonzheng
可以看到CMD就是ENTRYPIOINT的参数,CMD可以被覆盖,我们可以把命令相同的部分打入到ENTRYPIOINT,不同的部分通过CMD覆盖
ENV:配置环境变量
[root@k8s-master01 dockerfiles]# vim Dockerfile
# 修改为以下内容
FROM centos:8
LABEL maintainer="test dockerfile"
LABEL test=dockerfile
ENV test_env1 env1
ENV test_env2 env2
RUN useradd mingsonzheng
RUN mkdir /opt/mingsonzheng
#ENTRYPOINT ["echo"]
CMD echo "test_env1 test_env2"
#RUN useradd mingsonzheng && /opt/mingsonzheng
构建镜像
[root@k8s-master01 dockerfiles]# docker build -t centos:env .
启动容器
[root@k8s-master01 dockerfiles]# docker run -ti --rm centos:env
test_env1 test_env2
环境变量也可以这么写
ENV env1=test1 env2=test2
正式应用的ENV变量不要写在Dockerfile里面,应该使用k8s资源文件,或者docker的-e参数注入进去,这样也能减少构建镜像的层数
ADD:添加文件到容器
# 新增文件
[root@k8s-master01 dockerfiles]# echo 123 > index.html
# 压缩文件
[root@k8s-master01 dockerfiles]# tar zcf index.tar.gz index.html t/
# 编辑Dockerfile
[root@k8s-master01 dockerfiles]# vim Dockerfile
# 修改为以下内容
FROM centos:8
LABEL maintainer="test dockerfile"
LABEL test=dockerfile
ENV test_env1 env1
ENV test_env2 env2
RUN useradd mingsonzheng
RUN mkdir /opt/mingsonzheng
#ENTRYPOINT ["echo"]
ENV env1=test1 env2=test2
ADD ./index.tar.gz /opt/
CMD echo "test_env1 test_env2"
#RUN useradd mingsonzheng && /opt/mingsonzheng
构建镜像
[root@k8s-master01 dockerfiles]# docker build -t centos:add .
启动容器
[root@k8s-master01 dockerfiles]# docker run -ti --rm centos:add bash
[root@b2dacca0f34f /]# cd /opt/
[root@b2dacca0f34f opt]# ls
index.html mingsonzheng t
可以看到文件被复制到容器的指定目录下,并且自动解压了
COPY:复制文件到容器
[root@k8s-master01 dockerfiles]# vim Dockerfile
# 修改为以下内容
FROM centos:8
LABEL maintainer="test dockerfile"
LABEL test=dockerfile
ENV test_env1 env1
ENV test_env2 env2
RUN useradd mingsonzheng
RUN mkdir /opt/mingsonzheng
#ENTRYPOINT ["echo"]
ENV env1=test1 env2=test2
ADD ./index.tar.gz /opt/
COPY ./t /opt/mingsonzheng/
CMD echo "test_env1 test_env2"
#RUN useradd mingsonzheng && /opt/mingsonzheng
构建镜像
[root@k8s-master01 dockerfiles]# docker build -t centos:add .
启动容器
[root@k8s-master01 dockerfiles]# docker run -ti --rm centos:add bash
[root@3ca8cddbf8b3 /]# cd /opt/mingsonzheng/
[root@3ca8cddbf8b3 mingsonzheng]# ls
Dockerfile
可以看到docker把t目录下的文件复制进来了,并不会复制目录,如果想要复制目录的话,只能多加一层目录
COPY复制压缩文件的话不会解压
docker build的时候会把当前目录下所有东西发送到内存中,所以文件必须要放在执行docker build命令的文件目录下
WORKDIR:设置容器的工作目录
[root@k8s-master01 dockerfiles]# vim Dockerfile
# 修改为以下内容
FROM centos:8
LABEL maintainer="test dockerfile"
LABEL test=dockerfile
ENV test_env1 env1
ENV test_env2 env2
RUN useradd mingsonzheng
RUN mkdir /opt/mingsonzheng
#ENTRYPOINT ["echo"]
ENV env1=test1 env2=test2
ADD ./index.tar.gz /opt/
COPY ./t /opt/mingsonzheng/
CMD echo "test_env1 test_env2"
WORKDIR /opt/mingsonzheng
CMD pwd ; ls
#RUN useradd mingsonzheng && /opt/mingsonzheng
构建镜像
[root@k8s-master01 dockerfiles]# docker build -t centos:workdir .
启动容器
[root@k8s-master01 dockerfiles]# docker run -ti --rm centos:workdir
/opt/mingsonzheng
Dockerfile
USER:容器使用的用户
# 启动容器
[root@k8s-master01 dockerfiles]# docker run -ti --rm centos:workdir bash
# 查看当前用户
[root@87983e2581da mingsonzheng]# whoami
root
# 查看系统其他用户
[root@87983e2581da mingsonzheng]# cat /etc/passwd
mingsonzheng:x:1000:1000::/home/mingsonzheng:/bin/bash
切换用户
[root@k8s-master01 dockerfiles]# vim Dockerfile
# 修改为以下内容
FROM centos:8
LABEL maintainer="test dockerfile"
LABEL test=dockerfile
ENV test_env1 env1
ENV test_env2 env2
RUN useradd mingsonzheng
RUN mkdir /opt/mingsonzheng
#ENTRYPOINT ["echo"]
ENV env1=test1 env2=test2
ADD ./index.tar.gz /opt/
COPY ./t /opt/mingsonzheng/
CMD echo "test_env1 test_env2"
WORKDIR /opt/mingsonzheng
USER 1000
CMD pwd ; ls
#RUN useradd mingsonzheng && /opt/mingsonzheng
构建镜像
[root@k8s-master01 dockerfiles]# docker build -t centos:workdir .
启动容器
[root@k8s-master01 dockerfiles]# docker run -ti --rm centos:workdir bash
[mingsonzheng@f6110f7e8e3a mingsonzheng]$ whoami
mingsonzheng
使用镜像仓库地址(xxx.com)构建镜像
[root@k8s-master01 dockerfiles]# docker build -t xxx.com/centos:workdir .
查看镜像
[root@k8s-master01 dockerfiles]# docker images
REPOSITORY TAG IMAGE ID CREATED SIZE
xxx.com/centos workdir cf9572a3833e 3 minutes ago 210MB
VOLUME:创建挂载点
[root@k8s-master01 dockerfiles]# vim Dockerfile
# 修改为以下内容
FROM centos:8
LABEL maintainer="test dockerfile"
LABEL test=dockerfile
ENV test_env1 env1
ENV test_env2 env2
RUN useradd mingsonzheng
RUN mkdir /opt/mingsonzheng
#ENTRYPOINT ["echo"]
ENV env1=test1 env2=test2
ADD ./index.tar.gz /opt/
COPY ./t /opt/mingsonzheng/
CMD echo "test_env1 test_env2"
WORKDIR /opt/mingsonzheng
#USER 1000
VOLUME /data
CMD pwd ; ls
#RUN useradd mingsonzheng && /opt/mingsonzheng
构建镜像
[root@k8s-master01 dockerfiles]# docker build -t centos:workdir .
启动容器
[root@k8s-master01 dockerfiles]# docker run -ti --rm centos:workdir bash
[root@6aecced2b780 mingsonzheng]# cd /
[root@6aecced2b780 /]# ls
bin data dev etc home lib lib64 lost+found media mnt opt proc root run sbin srv sys tmp usr var
可以看到data目录,它会与本地 /var/lib/docker/volumes/ 的目录下生成一个volume的目录,然后挂载到容器中的data下,退出容器则会清除本地目录下的data
通过-v将mysql_data目录挂载到容器中,使用-v则系统不会自动创建挂载目录
[root@k8s-master01 ~]# mkdir mysql_data
[root@k8s-master01 ~]# docker run -ti -v /root/mysql_data/:/data xxx.com/centos:workdir bash
制作小镜像
一定不要使用centos基础镜像,可以使用alpine,busybox,scratch,debian
修改基础镜像
[root@k8s-master01 dockerfiles]# vim Dockerfile
# 修改为以下内容
#FROM centos:8
FROM alpine:3.8
LABEL maintainer="test dockerfile"
LABEL test=dockerfile
ENV test_env1 env1
ENV test_env2 env2
RUN adduser -D mingsonzheng
RUN mkdir -p /opt/mingsonzheng
#ENTRYPOINT ["echo"]
ENV env1=test1 env2=test2
ADD ./index.tar.gz /opt/
COPY ./t /opt/mingsonzheng/
WORKDIR /opt/mingsonzheng
#USER 1000
VOLUME /data
CMD pwd ; ls
#RUN useradd mingsonzheng && /opt/mingsonzheng
构建镜像
[root@k8s-master01 dockerfiles]# docker build -t centos:workdir .
查看镜像
[root@k8s-master01 dockerfiles]# docker images
REPOSITORY TAG IMAGE ID CREATED SIZE
alpine workdir 7fc05207414a 22 seconds ago 4.42MB
centos workdir b165153a0132 36 minutes ago 210MB
alpine与centos相比,SIZE非常小,适合做为基础镜像,基础镜像可以在官方仓库查找,不需要自己制作
如果需要用到glibc,可以使用node:slim,python:slim,net作为基础镜像
多阶段制作小镜像
分开两个步骤:编译操作和生成最终镜像的操作
新建一个go文件
[root@k8s-master01 dockerfiles]# vim main.go
# 修改为以下内容
package main
import "fmt"
func main() {
fmt.Println("Hello World")
}
新建一个Dockerfile文件
[root@k8s-master01 dockerfiles]# mv Dockerfile t/123
[root@k8s-master01 dockerfiles]# vim Dockerfile
# 修改为以下内容
FROM golang:1.14.4-alpine
WORKDIR /opt
COPY main.go /opt
RUN go build /opt/main.go
CMD "./main"
构建镜像
[root@k8s-master01 dockerfiles]# docker build -t hello:single_build .
启动容器
[root@k8s-master01 dockerfiles]# docker run -ti --rm hello:single_build
Hello World
查看镜像
[root@k8s-master01 dockerfiles]# docker images
REPOSITORY TAG IMAGE ID CREATED SIZE
hello single_build ca11178a0c13 44 seconds ago 372MB
查看包大小
[root@k8s-master01 dockerfiles]# docker run -ti --rm hello:single_build sh
/opt # ls -lh
total 2M
-rwxr-xr-x 1 root root 2.0M Jul 16 08:30 main
-rw-r--r-- 1 root root 77 Jul 16 08:24 main.go
/opt #
2M的包生成的镜像有372M,我们应该使用分阶段构建
[root@k8s-master01 dockerfiles]# vim Dockerfile
# 修改为以下内容
# build step
FROM golang:1.14.4-alpine
WORKDIR /opt
COPY main.go /opt
RUN go build /opt/main.go
CMD "./main"
# create real app image
FROM alpine:3.8
COPY --from=0 /opt/main /
CMD "./opt/main"
--from=0就是第一步的操作(# create real app image上面的操作)
构建镜像
[root@k8s-master01 dockerfiles]# docker build -t hello:alpine .
启动容器
[root@k8s-master01 dockerfiles]# docker run -ti --rm hello:alpine sh
/ # ls
bin dev etc home lib main media mnt proc root run sbin srv sys tmp usr var
/ # ./main
Hello World
查看镜像
[root@k8s-master01 dockerfiles]# docker images
REPOSITORY TAG IMAGE ID CREATED SIZE
hello alpine 66fa14e35f3c About a minute ago 6.48MB
hello single_build ca11178a0c13 11 minutes ago 372MB
可以看到单阶段和多阶段镜像大小的区别,所以一定要进行分阶段构建
使用builder命令复制包到根目录
[root@k8s-master01 dockerfiles]# vim Dockerfile
# 修改为以下内容
# build step
FROM golang:1.14.4-alpine as builder
WORKDIR /opt
COPY main.go /opt
RUN go build /opt/main.go
CMD "./main"
# create real app image
FROM alpine:3.8
COPY --from=builder /opt/main /
CMD "./opt/main"
构建镜像
[root@k8s-master01 dockerfiles]# docker build -t hello:alpine .
查看镜像
[root@k8s-master01 dockerfiles]# docker images
REPOSITORY TAG IMAGE ID CREATED SIZE
hello alpine 66fa14e35f3c 9 minutes ago 6.48MB
镜像大小和前面的差别不大,我们只是将第一个步骤打包成镜像,然后把它里面的文件复制到现在的镜像里面
php Dockerfile
[root@k8s-master01 dockerfiles]# mv Dockerfile goDockerfile
[root@k8s-master01 dockerfiles]# vim Dockerfile
# 修改为以下内容
FROM php:7.1.22-fpm-alpine
RUN apk add --no-cache binutils freetype libpng libjpeg-turbo freetype-dev libpng-dev libjpeg-turbo-dev libc6-compat libxml2 libxml2-dev libmcrypt libmcrypt-dev libc-dev icu-dev gettext-dev openssl-dev bzip2-dev
RUN docker-php-ext-install pdo pdo_mysql mcrypt zip gd pcntl opcache bcmath
#RUN docker-php-ext-install gettext
RUN docker-php-ext-install mysqli
#RUN apk add --no-cache php7-sysvsem php7-pdo_dblib php7-sockets php-soap php7-xmlrpc
##RUN apk add --no-cache php7-sysvsem php7-pdo_dblib php7-sockets php-soap php7-xmlrpc
##RUN apk add --no-cache freetds-dev
##RUN docker-php-ext-install pdo_dblib
#RUN docker-php-ext-install soap
#RUN docker-php-ext-install sockets
#RUN docker-php-ext-install sysvsem
#RUN docker-php-ext-install xmlrpc
#RUN apk add --no-cache freetds-dev
#RUN docker-php-ext-install pdo_dblib
#RUN docker-php-ext-configure gd --with-freetype-dir=/usr/include/ --with-jpeg-dir=/usr/include/
#RUN docker-php-ext-install -j$(nproc) gd
#FROM php:7.1.22-fpm-alpine
#COPY --from=0 /usr/local/lib/php/extensions/no-debug-non-zts-20160303 /usr/local/lib/php/extensions/no-debug-non-zts-20160303
#RUN apk add --no-cache freetds-dev php7-sysvsem php7-pdo_dblib php7-sockets php-soap php7-xmlrpc binutils freetype libpng libjpeg-turbo freetype-dev libpng-dev libjpeg-turbo-dev libc6-compat libxml2 libxml2-dev libmcrypt libmcrypt-dev libc-dev icu-dev gettext-dev openssl-dev bzip2-dev && cd /usr/local/lib/php/extensions/no-debug-non-zts-20160303 && docker-php-ext-enable *.so && rm -rf /var/cache/apk/*
- 第一步是选择基础镜像php:7.1.22-fpm-alpine
- 第二步是根据基础镜像安装依赖环境,php的一堆扩展包
多阶段的就是编译在一个镜像里面完成,第二个镜像使用第一个镜像编译的产物,省去了编译产生的缓存
Scratch空镜像
Scratch空镜像不可拉取,但是可以直接使用
使用scratch
[root@k8s-master01 dockerfiles]# cp Dockerfile t/phpDockerfile
[root@k8s-master01 dockerfiles]# cp goDockerfile Dockerfile
cp: overwrite ‘Dockerfile’? y
[root@k8s-master01 dockerfiles]# vim Dockerfile
# 修改为以下内容
# build step
FROM golang:1.14.4-alpine as builder
WORKDIR /opt
COPY main.go /opt
RUN go build /opt/main.go
CMD "./main"
# create real app image
FROM scratch
COPY --from=builder /opt/main /
CMD "./main"
构建镜像
[root@k8s-master01 dockerfiles]# docker build -t hello:scratch .
查看镜像
[root@k8s-master01 dockerfiles]# docker images
REPOSITORY TAG IMAGE ID CREATED SIZE
hello scratch 4d747be3b21b About a minute ago 2.07MB
启动容器
[root@k8s-master01 dockerfiles]# docker run -ti --rm hello:scratch /main
Hello World
基于最基础的镜像,制作依赖环境的镜像可以使用,但是建议使用官方制作好的镜像
如果Dockerfile中引用的镜像发生了更新,需要使用pull参数