顶级 services 配置

一、基本配置

vim docker-compose.yml

version: "3.8"
services:
  h1:
    image: "nginx"
    container_name: docker-myhosts

[root@Docker1 compose]# docker-compose up -d
[root@Docker1 compose]# docker-compose ps

     Name                   Command               State   Ports 
----------------------------------------------------------------

docker-myhosts   /docker-entrypoint.sh ngin ...   Up      80/tcp

二、Build

a.简单使用

[root@Docker-1 compose]# cat docker-compose.yml 
version: "3.8"
services:
  webapp:
    build: .
[root@Docker-1 compose]# cat Dockerfile 
FROM nginx
COPY ./index.html /usr/share/nginx/html/index.html
[root@Docker-1 compose]# echo Hello,neko > index.html
[root@Docker-1 compose]# ls
docker-compose.yml  Dockerfile  index.html

[root@Docker-1 compose]# docker-compose up -d
WARNING: Found orphan containers (docker-myhosts, host1) for this project. If you removed or renamed this service in your compose file, you can run this command with the --remove-orphans flag to clean it up.
Building webapp
Step 1/2 : FROM nginx
 ---> f6d0b4767a6c
Step 2/2 : COPY ./index.html /usr/share/nginx/html/index.html
 ---> 236d0b8f3cf2

Successfully built 236d0b8f3cf2
Successfully tagged compose_webapp:latest
WARNING: Image for service webapp was built because it did not already exist. To rebuild this image you must use `docker-compose build` or `docker-compose up --build`.
Creating compose_webapp_1 ... done

[root@Docker-1 compose]# docker images compose_webapp:latest 
REPOSITORY       TAG       IMAGE ID       CREATED          SIZE
compose_webapp   latest    236d0b8f3cf2   57 seconds ago   133MB
[root@Docker-1 compose]# docker-compose ps

      Name                    Command               State   Ports 
------------------------------------------------------------------

compose_webapp_1   /docker-entrypoint.sh ngin ...   Up      80/tcp

b.构建lnmp环境

[root@Docker-1 neko]# mkdir mysql nginx php
[root@Docker-1 neko]# ls
docker-compose.yml  Dockerfile  mysql  nginx  php
[root@Docker-1 neko]# cp Dockerfile nginx/
[root@Docker-1 neko]# cp Dockerfile mysql/
[root@Docker-1 neko]# cp Dockerfile php/
[root@Docker-1 neko]# tree .
.
├── docker-compose.yml
├── Dockerfile
├── mysql
│   └── Dockerfile
├── nginx
│   ├── Dockerfile
│   └── index.html
└── php
    └── Dockerfile

[root@Docker-1 neko]# cat docker-compose.yml 
version: "3.8"
services:
  webapp:
    build: ./nginx
  db:
    build: ./mysql
  php:
    build: ./php

销毁之前的compose

docker-compose down

[root@Docker-1 neko]# cat nginx/Dockerfile 
FROM nginx
COPY ./index.html /usr/share/nginx/html/index.html

[root@Docker-1 neko]# cat mysql/Dockerfile 
FROM mysql

[root@Docker-1 neko]# cat php/Dockerfile
FROM php:7.4-fpm

构建 不启动

docker-compose build

查看项目

[root@Docker-1 neko]# docker images |grep neko
neko_webapp                                     latest         236d0b8f3cf2   35 minutes ago   133MB
neko_db                                         latest         c8562eaf9d81   2 days ago       546MB
neko_php                                        latest         f5460fa2369d   9 days ago       405MB

报错解决

[root@Docker-1 compose]# docker-compose up -d
WARNING: Found orphan containers (myhost) for this project. If you removed or renamed this service in your compose file, you can run this command with the --remove-orphans flag to clean it up.

解决办法：

docker-compose up -d --remove-orphans

三、Image

指定为镜像名称或镜像 ID。如果镜像在本地不存在，Compose 将会尝试拉取这个镜像。

image: ubuntu
image: orchardup/postgresql
image: a4bc65fd

[root@Docker-1 ~]# mkdir image-compose
[root@Docker-1 ~]# cd image-compose/
[root@Docker-1 image-compose]# cat docker-compose.yml 
version: "3.8"
services:
  neko-1:
    image: nginx
[root@Docker-1 image-compose]# docker-compose up -d
Creating network "image-compose_default" with the default driver
Creating image-compose_neko-1_1 ... done
[root@Docker-1 image-compose]# docker-compose  ps

         Name                       Command               State   Ports 
------------------------------------------------------------------------

image-compose_neko-1_1   /docker-entrypoint.sh ngin ...   Up      80/tcp

如果同时指定了 image和 build， image 不在具有单独使用它的意义，而是指定了目前要构建的镜像的名称。也就是说 Compose 会使用 build 指令中指定的 Dockerfilel 构建的镜像，之后构建的镜像名称使用 image 中指定的名字 webapp:tag命名。

[root@Docker-1 image-compose]# cat Dockerfile
FROM nginx
[root@Docker-1 image-compose]# vim docker-compose.yml 
[root@Docker-1 image-compose]# cat docker-compose.yml 
version: "3.8"
services:
  neko-1:
    image: nginx
  neko-2:
    build: .
    image: hello.c
[root@Docker-1 image-compose]# docker-compose up -d
image-compose_neko-1_1 is up-to-date
Creating image-compose_neko-2_1 ... done
[root@Docker-1 image-compose]# 

四、Command

使用 command 可以覆盖容器启动后默认执行的命令。

command: bundle exec thin -p 3000

也可以写成类似 Dockerfile 中的格式：

command: [bundle, exec, thin, -p, 3000]

[root@Docker-1 ~]# mkdir command
[root@Docker-1 ~]# cd command/
[root@Docker-1 command]# cat docker-compose.yml 
version: "3"
services:
  app-1:
    image: nginx
    command: /bin/sh
    container_name: nginx-sh
  app-2:
    image: nginx
    container_name: nginx-off


[root@Docker-1 command]# docker-compose up -d
Creating network "command_default" with the default driver
Creating nginx-sh  ... done
Creating nginx-off ... done
[root@Docker-1 command]# docker-compose ps 

  Name                 Command               State    Ports 
------------------------------------------------------------

nginx-off   /docker-entrypoint.sh ngin ...   Up       80/tcp
nginx-sh    /docker-entrypoint.sh /bin/sh    Exit 0 

五、Depends_on

解决容器的依赖、启动先后的问题。以下例子中会先启动容器 db 和 redis 再启动 web

version: '3'

services:
  web:
    build: .
    depends_on:
      - db
      - redis

  redis:
    image: redis

  db:
    image: postgres

注意：web 服务不会等待 redis 和 db 「完全启动」之后才启动。

六、Environment

[root@Docker-1 ~]# mkdir env
[root@Docker-1 ~]# cd env/
[root@Docker-1 env]# cat docker-compose.yml
version: "3.8"
service:
  h1:
    image: alpine
    stdin_open: true # docker run -i
    tty: true        # docker run -i
    environment:
      user: neko
      TEST:
[root@Docker-1 env]# docker run nginx env
HOSTNAME=0e58b69d59dc
HOME=/root
PKG_RELEASE=1~buster
NGINX_VERSION=1.19.6
PATH=/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin
NJS_VERSION=0.5.0
PWD=/

七、Expose 和 Ports

暴露端口 expose

暴露端口，但不映射到宿主机，只被其他容器连接到此容器时使用。
仅可以指定容器内部的端口为参数

映射端口 ports

使用：
宿主端口：容器端口 (即：HOST:CONTAINER) 的格式格式，或者仅仅指定容器的端口（宿主将会随机选择端口）都可以。

[root@Docker-1 ~]# mkdir ports
[root@Docker-1 ~]# cd ports/
[root@Docker-1 ports]# cat docker-compose.yml 
version: "3.8"
services:
  web1:
    image: nginx
    ports:
      - "9000:80"  # 相当于 docker run -p 9000:80

[root@Docker-1 ports]# docker-compose up -d
Creating network "ports_default" with the default driver
Creating ports_web1_1 ... done
[root@Docker-1 ports]# docker-compose ps

    Name                  Command               State          Ports        
----------------------------------------------------------------------------

ports_web1_1   /docker-entrypoint.sh ngin ...   Up      0.0.0.0:9000->80/tcp

然后访问端口

curl 192.168.31.248:9000  #访问宿主机9000端口

八、Extra_hosts

[root@Docker-1 ~]# mkdir extra_hosts
[root@Docker-1 ~]# cd extra_hosts/
[root@Docker-1 extra_hosts]# cat docker-compose.yml 
version: "3.8"
services:
  web1:
    image: nginx
    ports:
      - "9010:80"  # 相当于 docker run -p 9010:80
    extra_hosts:
      - "web2:172.18.0.9"
      - "www.nekoosu.com:1.2.3.4"

[root@Docker-1 extra_hosts]# docker-compose up -d
Creating extra_hosts_web1_1 ... done

进去系统发现 已添加hosts

[root@Docker-1 extra_hosts]# docker-compose exec web1 sh

# cat /etc/hosts

127.0.0.1       localhost
::1     localhost ip6-localhost ip6-loopback
fe00::0 ip6-localnet
ff00::0 ip6-mcastprefix
ff02::1 ip6-allnodes
ff02::2 ip6-allrouters
172.18.0.9      web2
1.2.3.4 www.nekoosu.com
172.28.0.2      ddc53752333e

九、Networks

[root@Docker-1 ~]# mkdir network
[root@Docker-1 ~]# cd network/
[root@Docker-1 network]# cat docker-compose.yml
version: "3.8"
services:
  web1:
    image: nginx
    ports:
      - "9020:80"  # 相当于 docker run -p 9020:80
    networks:
      - web-net # 自定义网络的名称

  web2:
    image: centos
    tty: true
    stdin_open: true
    networks:
      - web-net # docker run --network=web-net
    
networks:  # docker network create
  web-net:
  
  
[root@Docker-1 network]# docker-compose up -d
Creating network "network_web-net" with the default driver
Pulling web2 (centos:)...
latest: Pulling from library/centos
7a0437f04f83: Already exists
Digest: sha256:5528e8b1b1719d34604c87e11dcd1c0a20bedf46e83b5632cdeac91b8c04efc1
Status: Downloaded newer image for centos:latest
Creating network_web1_1 ... done
Creating network_web2_1 ... done

查看是否有自定义的网络

[root@Docker-1 network]# docker network inspect network_web-net
        "Containers": {
    
    
            "c12d9c084743248e0f1c8c85d651f2c795ceec4ba3c0365ecd927c6ccaef3241": {
    
    
                "Name": "network_web1_1",
                "EndpointID": "063b66b7bba0346a9b2cab71958ed9b52083eab64c35e28476202ad4dd46d4c8",
                "MacAddress": "02:42:ac:1d:00:02",
                "IPv4Address": "172.29.0.2/16",
                "IPv6Address": ""
            },
            "d325643ae6ae53597370a3b835f03ed786a4f918dfa4608aac940c7491791e7f": {
    
    
                "Name": "network_web2_1",
                "EndpointID": "a2feb44407bae619ed5d082e5eaf3a20c987c0bd616ab00d39ac351c8044353a",
                "MacAddress": "02:42:ac:1d:00:03",
                "IPv4Address": "172.29.0.3/16",
                "IPv6Address": ""

查看是否通信

[root@Docker-1 network]# docker-compose exec web2 bash
[root@d325643ae6ae /]# ls
bin  etc   lib    lost+found  mnt  proc  run   srv  tmp  var
dev  home  lib64  media       opt  root  sbin  sys  usr
[root@d325643ae6ae /]# ping web1
PING web1 (172.29.0.2) 56(84) bytes of data.
64 bytes from network_web1_1.network_web-net (172.29.0.2): icmp_seq=1 ttl=64 time=0.073 ms
^C
--- web1 ping statistics ---
1 packets transmitted, 1 received, 0% packet loss, time 0ms
rtt min/avg/max/mdev = 0.073/0.073/0.073/0.000 ms
[root@d325643ae6ae /]# curl web1
<!DOCTYPE html>
<html>
<head>
<title>Welcome to nginx!</title
...


[root@Docker-1 ~]# mkdir web-db-redis
[root@Docker-1 ~]# cd web-db-redis/
[root@Docker-1 web-db-redis]# cat docker-compose.yml
version: "3"
services:
  web1:
    image: centos
    stdin_open: true
    tty: true
    ports:

   - "9030:80"
     tworks:
        - net-web
          net-db
            db:
              image: centos
              stdin_open: true
              tty: true
              networks:
             - net-db
               redis:
                   image: redis:alpine
                   networks:
                  - net-db
                    networks:
                      net-db:
                      net-web:

[root@Docker-1 web-db-redis]# docker-compose up -d
Creating network "web-db-redis_net-web" with the default driver
Creating network "web-db-redis_net-db" with the default driver
Creating web-db-redis_web1_1  ... done
Creating web-db-redis_redis_1 ... done
Creating web-db-redis_db_1    ... done
[root@Docker-1 web-db-redis]# docker-compose ps

        Name                      Command               State           Ports        
------------------------------------------------------------------------------------

web-db-redis_db_1      /bin/bash                        Up                          
web-db-redis_redis_1   docker-entrypoint.sh redis ...   Up      6379/tcp            
web-db-redis_web1_1    /bin/bash                        Up      0.0.0.0:9030->80/tcp

检测网络连通性

[root@Docker-1 web-db-redis]# docker-compose exec web1 bash
[root@f416f6829256 /]# ping redis
PING redis (192.168.32.2) 56(84) bytes of data.
64 bytes from web-db-redis_redis_1.web-db-redis_net-db (192.168.32.2): icmp_seq=1 ttl=64 time=0.167 ms
64 bytes from web-db-redis_redis_1.web-db-redis_net-db (192.168.32.2): icmp_seq=2 ttl=64 time=0.051 ms
^C
--- redis ping statistics ---
2 packets transmitted, 2 received, 0% packet loss, time 3ms
rtt min/avg/max/mdev = 0.051/0.109/0.167/0.058 ms
[root@f416f6829256 /]# ping db   
PING db (192.168.32.3) 56(84) bytes of data.
64 bytes from web-db-redis_db_1.web-db-redis_net-db (192.168.32.3): icmp_seq=1 ttl=64 time=0.065 ms
64 bytes from web-db-redis_db_1.web-db-redis_net-db (192.168.32.3): icmp_seq=2 ttl=64 time=0.051 ms
^C
--- db ping statistics ---
2 packets transmitted, 2 received, 0% packet loss, time 3ms
rtt min/avg/max/mdev = 0.051/0.058/0.065/0.007 ms
[root@f416f6829256 /]# ping web1
PING web1 (192.168.32.4) 56(84) bytes of data.
64 bytes from f416f6829256 (192.168.32.4): icmp_seq=1 ttl=64 time=0.035 ms
64 bytes from f416f6829256 (192.168.32.4): icmp_seq=2 ttl=64 time=0.031 ms
^C
--- web1 ping statistics ---
2 packets transmitted, 2 received, 0% packet loss, time 999ms
rtt min/avg/max/mdev = 0.031/0.033/0.035/0.002 ms

1.指定 ip

基于之前的操作

[root@Docker-1 web-db-redis]# cd
[root@Docker-1 ~]# cd network/
[root@Docker-1 network]# docker-compose ps
Name                   Command               State          Ports       
------------------------------------------------------------------------------
network_web1_1   /docker-entrypoint.sh ngin ...   Up      0.0.0.0:9020->80/tcp
network_web2_1   /bin/bash                        Up                          
[root@Docker-1 network]# docker-compose exec web1 hostname -i
172.29.0.2
[root@Docker-1 network]# docker-compose exec web2 hostname -i
172.29.0.3

指定 ipv4 和 ipv6

[root@Docker-1 network]# cat docker-compose.yml
version: "3.8"
services:
  web1:
    image: 
    tty: true
    stdin_open: true

    ports:
      - "9020:80"  # 相当于 docker run -p 9020:80
    networks:
      web-net: # 自定义网络的名称
        ipv4_address: 172.16.238.10
        ipv6_address: 2001:3984:3989::10

  web2:
    image: centos
    tty: true
    stdin_open: true
    networks:
      - web-net # docker run --network=web-net
    
networks:  # docker network create
  web-net:  # 名称
    driver: bridge  
    enable_ipv6: true   # 是否用 ipv6 是 true 否 force
    ipam:
      driver: default   # 网络模式
      config:
        - subnet: 172.16.238.0/24  # 指定网络号  可以修改 与ipv4 对应
          gateway: 172.16.238.1
        - subnet: 2001:3984:3989::/64
          gateway: 2001:3984:3989::1

重新创建网络

[root@Docker-1 network]# docker-compose down
Stopping network_web2_1 ... done
Stopping network_web1_1 ... done
Removing network_web2_1 ... done
Removing network_web1_1 ... done
Removing network network_web-net
[root@Docker-1 network]# docker-compose up -d
Creating network "network_web-net" with driver "bridge"
Creating network_web2_1 ... done
Creating network_web1_1 ... done
[root@Docker-1 network]#  docker-compose exec web1 ip a
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
       valid_lft forever preferred_lft forever
    inet6 ::1/128 scope host 
       valid_lft forever preferred_lft forever
60: eth0@if61: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default 
    link/ether 02:42:ac:10:ee:0a brd ff:ff:ff:ff:ff:ff link-netnsid 0
    inet 172.16.238.10/24 brd 172.16.238.255 scope global eth0
       valid_lft forever preferred_lft forever
    inet6 2001:3984:3989::10/64 scope global nodad 
       valid_lft forever preferred_lft forever
    inet6 fe80::42:acff:fe10:ee0a/64 scope link 
       valid_lft forever preferred_lft forever
[root@Docker-1 network]#  docker-compose exec web2 ip a
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
       valid_lft forever preferred_lft forever
    inet6 ::1/128 scope host 
       valid_lft forever preferred_lft forever
54: eth0@if55: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default 
    link/ether 02:42:ac:10:ee:02 brd ff:ff:ff:ff:ff:ff link-netnsid 0
    inet 172.16.238.2/24 brd 172.16.238.255 scope global eth0
       valid_lft forever preferred_lft forever
    inet6 2001:3984:3989::2/64 scope global nodad 
       valid_lft forever preferred_lft forever
    inet6 fe80::42:acff:fe10:ee02/64 scope link 
       valid_lft forever preferred_lft forever

2.使用现有网络

如果您希望容器加入现有网络，请使用以下external 选项.

version: "3"
servers:
   host1:
      image: centos
      stdin_open: true
      tty: true
networks:

  # 此项目中的服务使用的默认网络，此网络不会被创建

  # 实际会使用 my-pre-existing-network

  default:
    external:
      name: my-pre-existing-network

这种情况下，Compose不会尝试创建名为 [projectname]_default 的网络，而是查找名为 my-pre-existing-network 的网络，并将应用程序的容器连接到该网络。

也可以把默认 default 网络改成其他的名称，比如
version: "3.8"
services:
   host1:
     image: centos
     stdin_open: true
     tty: true
     networks:
       neko:
networks:
  neko:
    external:
      name: my-pre-existing-network

十、Aliases

网络上此服务的别名 备用名称

格式：

version: "3"
services:
  host1:
    image: centos
    expose:
      - "4500"
    stdin_open: true
    networks:
      network-1:
        aliases:
         - h1
      network-2:
        aliases:
         - h2
  host2:
    image: centos
    stdin_open: true
    networks:
        network-1:
  host3:
      image: centos
      stdin_open: true
      networks:
        network-2:
networks:
  h1
  h2


[root@docker-host aliases]# docker-compose ps
    Name        Command    State    Ports  
-------------------------------------------
aliases_host1_1   /bin/bash   Up      4500/tcp
aliases_host2_1   /bin/bash   Up              
aliases_host3_1   /bin/bash   Up 

十一、Healthcheck

通过命令检查容器是否健康运行。

[root@Docker-1 ~]# mkdir healthcheck
[root@Docker-1 ~]# cd healthcheck/

[root@Docker-1 healthcheck]# cat docker-compose.yml
version: "3"
services:
  web1:
    image: nginx
    healthcheck:
      test: ["CMD", "curl", "-f", "http://localhost"]
      interval: 1m30s
      timeout: 10s
      retries: 3
[root@Docker-1 healthcheck]# docker-compose up -d
Creating network "healthcheck_default" with the default driver
Creating healthcheck_web1_1 ... done
[root@Docker-1 healthcheck]# docker-compose ps

       Name                     Command                       State           Ports 
------------------------------------------------------------------------------------

healthcheck_web1_1   /docker-entrypoint.sh ngin ...   Up (health: starting)   80/tcp

等到达设定时间 1分30秒 后

[root@Docker-1 healthcheck]# docker-compose ps

       Name                     Command                  State       Ports 
---------------------------------------------------------------------------

healthcheck_web1_1   /docker-entrypoint.sh ngin ...   Up (healthy)   80/tcp

拓扑：

删除Linux文件中的 中文字符

ls | grep -vP '[a-z0-9]' | rm -rf `xargs`