1.docker compose安装
#自动 sudo yum update sudo yum install docker-compose-plugin docker compose version #手动安装 DOCKER_CONFIG=${DOCKER_CONFIG:-$HOME/.docker} mkdir -p $DOCKER_CONFIG/cli-plugins curl -SL https://github.com/docker/compose/releases/download/v2.11.1/docker-compose-linux-x86_64 -o $DOCKER_CONFIG/cli-plugins/docker-compose 此命令下载最新版本的 Docker Compose（从 Compose 发布存储库）并在$HOME目录下为活动用户安装 Compose。 #安装： Docker Compose 为您系统上的所有用户，替换~/.docker/cli-plugins为/usr/local/lib/docker/cli-plugins. 不同版本的 Compose，替换v2.11.1为您要使用的 Compose 版本。 #对二进制文件应用可执行权限： chmod +x $DOCKER_CONFIG/cli-plugins/docker-compose #或者，如果您选择为所有用户安装 Compose： sudo chmod +x /usr/local/lib/docker/cli-plugins/docker-compose #测试安装 docker compose version
2.docker compose配置文件编写 [root@localhost shell]# cat docker-compose-test.yml version: '3.9' services: #elastic es: image: 172.16.4.17:8090/public/elastic:7.1.1 restart: always deploy: resources: limits: memory: '4G' reservations: memory: '3G' container_name: es environment: - TZ=Asia/Shanghai - ELASTIC_PASSWORD=ytx@1234 volumes: - type: bind source: /etc/localtime target: /etc/localtime:ro - type: bind source: /data/aibox-common/elastic/data target: /usr/share/elasticsearch/data - type: bind source: /data/aibox-common/elastic/config target: /usr/share/elasticsearch/config ports: - 9200:9200 - 9300:9300 network_mode: "host" #kibanan kibanan: image: 172.16.4.17:8090/public/kibana:7.1.1 restart: always deploy: resources: limits: memory: '2G' reservations: memory: '1G' container_name: kibanan environment: - TZ=Asia/Shanghai volumes: - type: bind source: /etc/localtime target: /etc/localtime:ro - type: bind source: /data/aibox-common/kibana/config/kibana.yml target: /usr/share/kibana/config/kibana.yml ports: - 5601:5601 network_mode: "host" #mysql mysql: image: 172.16.4.17:8090/public/mysql:5.7.29 restart: always deploy: resources: limits: memory: '3G' reservations: memory: '2G' container_name: mysql environment: - TZ=Asia/Shanghai - MYSQL_ROOT_PASSWORD=ytx@1234 volumes: - type: bind source: /etc/localtime target: /etc/localtime:ro - type: bind source: /data/aibox-common/mysql/config/my.cnf target: /etc/mysql/my.cnf - type: bind source: /data/aibox-common/mysql/datadir target: /var/lib/mysql ports: - 3306:3306 network_mode: "host" #redis redis: image: 172.16.4.17:8090/public/redis:6.0 restart: always deploy: resources: limits: memory: '3G' reservations: memory: '2G' container_name: redis volumes: - type: bind source: /etc/localtime target: /etc/localtime:ro - type: bind source: /data/aibox-common/redis/conf/redis.conf target: /usr/local/etc/redis/redis.conf ports: - 6379:6379 command: redis-server /usr/local/etc/redis/redis.conf network_mode: "host" #nginx nginx: image: 172.16.4.17:8090/public/nginx:v1.20.1 restart: always container_name: nginx volumes: - type: bind source: /etc/localtime target: /etc/localtime:ro - type: bind source: /data/service/nginx/conf target: /data/service/nginx/conf - type: bind source: /data/service/nginx/logs target: /data/service/nginx/logs - type: bind source: /data/service/nginx/html target: /data/service/nginx/html ports: - 80:80 - 28882:28882 network_mode: "host" #device-server device-server: image: #镜像地址 container_name: device-server restart: always environment: - TZ=Asia/Shanghai volumes: - type: bind source: /etc/localtime target: /etc/localtime:ro - type: bind source: /data/aibox-common/device-server/config target: /device-server/config - type: bind source: /data/aibox-common/device-server/logs target: /device-server/logs - type: bind source: /data/pictures/ target: /device-server/pic/ ports: - 24444:24444 - 23333:23333 depends_on: - mysql - redis - es network_mode: "host" #cm-server cm-server: image: #镜像地址 container_name: cm-server restart: always environment: - TZ=Asia/Shanghai volumes: - type: bind source: /etc/localtime target: /etc/localtime:ro - type: bind source: /data/aibox-common/cm-server/config target: /aibox-control-manage/config - type: bind source: /data/aibox-common/cm-server/logs target: /aibox-control-manage/logs - type: bind source: /data/pictures/ target: /aibox-control-manage/pic - type: bind source: /data/aibox-common/cm-server/rlface_cpu target: /aibox-control-manage/rlface_cpu ports: - 28877:28877 depends_on: - mysql - redis - es - device-server network_mode: "host" #device-agent device-agent: image: #镜像地址 container_name: device-agent restart: always environment: - TZ=Asia/Shanghai volumes: - type: bind source: /etc/localtime target: /etc/localtime:ro - type: bind source: /data/aibox-common/ai-server target: /home/nvidia/aibox/ai-server - type: bind source: /data/aibox-common/device-agent/logs target: /home/nvidia/aibox/logs - type: bind source: /data/aibox-common/aimodel target: /home/nvidia/aibox/aimodel - type: bind source: /data/aibox-common/common target: /home/nvidia/aibox/common ports: - 28888:28888 depends_on: - mysql - redis - es - device-server - cm-server network_mode: "host" #ai-server ai-server: image: #镜像地址 container_name: ai-server restart: always environment: - TZ=Asia/Shanghai volumes: - type: bind source: /etc/localtime target: /etc/localtime:ro - type: bind source: /data/aibox-common/ai-server/logs target: /home/nvidia/aibox/logs - type: bind source: /data/aibox-common/aimodel target: /home/nvidia/aibox/aimodel - type: bind source: /data/aibox-common/common target: /home/nvidia/aibox/common ports: - 28889:28889 depends_on: - mysql - redis - es - device-server - cm-server - device-agent ulimits: core: 0 deploy: resources: reservations: devices: - capabilities: [gpu] #使用所有gpu network_mode: "host"
3.docker compose GPU 配置
ai-server: image: #镜像地址 container_name: ai-server restart: always environment: - TZ=Asia/Shanghai volumes: - type: bind source: /etc/localtime target: /etc/localtime:ro - type: bind source: /data/aibox-common/ai-server/logs target: /home/nvidia/aibox/logs - type: bind source: /data/aibox-common/aimodel target: /home/nvidia/aibox/aimodel - type: bind source: /data/aibox-common/common target: /home/nvidia/aibox/common ports: - 28889:28889 depends_on: - mysql - redis - es - device-server - cm-server - device-agent ulimits: core: 0 deploy: resources: reservations: devices: - driver: nvidia device_ids: ['0', '3'] #使用0、3号gpu capabilities: [gpu] network_mode: "host"
4.docker compose常用命令 0.官方文档：https://docs.docker.com/compose/reference/overview/ docker-compose [-f <arg>...] [options] [COMMAND] [ARGS...] 部分命令选项如下： -f，–file：指定使用的 Compose 模板文件，默认为 docker-compose.yml，可以多次指定，指定多个 yml； -p, --project-name：指定工程名称，默认使用 docker-compose.yml 文件所在目录的名称； -v：打印版本并退出； –log-level：定义日志等级（DEBUG, INFO, WARNING, ERROR, CRITICAL）。 1.help docker-compose -help查看帮助。 2.config docker-compose config -q验证docker-compose.yml文件。当配置正确时，不输出任何内容，当配置错误时，输出错误信息。 3.pull docker-compose pull拉取服务依赖的镜像。 # 拉取工程中所有服务依赖的镜像 docker-compose pull # 拉取工程中 nginx 服务依赖的镜像 docker-compose pull nginx # 拉取镜像过程中不打印拉取进度信息 docker-compose pull -q 4.up docker-compose up创建并启动所有服务的容器。指定多个yml加-f选项。以守护进程模式运行加-d选项。 # 前台启动 docker-compose up # 后台启动 docker-compose up -d # -f 指定使用的 Compose 模板文件，默认为 docker-compose.yml，可以多次指定，指定多个 yml docker-compose -f docker-compose.yml up -d 5.logs docker-compose logs查看服务容器的输出日志。默认情况下，docker-compose将对不同的服务输出使用不同的颜色来区分。可以通过--no-color来关闭颜色。 # 输出日志，不同的服务输出使用不同的颜色来区分 docker-compose logs # 跟踪日志输出 docker-compose logs -f # 关闭颜色 docker-compose logs --no-color 6.ps docker-compose ps列出工程中所有服务的容器。 # 列出工程中所有服务的容器 docker-compose ps # 列出工程中指定服务的容器 docker-compose ps nginx 7.run docker-compose run在指定服务容器上执行一个命令。 # 在工程中指定服务的容器上执行 echo "helloworld" docker-compose run nginx echo "helloworld" 8.exec docker-compose exec进入服务容器。 # 进入工程中指定服务的容器 docker-compose exec nginx bash # 当一个服务拥有多个容器时，可通过 --index 参数进入到该服务下的任何容器 docker-compose exec --index=1 nginx bash 9.pause docker-compose pause暂停服务容器 # 暂停工程中所有服务的容器 docker-compose pause # 暂停工程中指定服务的容器 docker-compose pause nginx 10.unpause docker-compose unpause恢复服务容器。 # 恢复工程中所有服务的容器 docker-compose unpause # 恢复工程中指定服务的容器 docker-compose unpause nginx 11.restart docker-compose restart重启服务容器。 # 重启工程中所有服务的容器 docker-compose restart # 重启工程中指定服务的容器 docker-compose restart nginx 12.start docker-compose start启动服务容器。 # 启动工程中所有服务的容器 docker-compose start # 启动工程中指定服务的容器 docker-compose start nginx 13.stop docker-compose stop停止服务容器。 # 停止工程中所有服务的容器 docker-compose stop # 停止工程中指定服务的容器 docker-compose stop nginx 14.kill docker-compose kill通过发送SIGKILL信号停止指定服务的容器。 # 通过发送 SIGKILL 信号停止工程中指定服务的容器 docker-compose kill nginx 15.rm docker-compose rm 删除服务（停止状态）容器。 # 删除所有（停止状态）服务的容器 docker-compose rm # 先停止所有服务的容器，再删除所有服务的容器 docker-compose rm -s # 不询问是否删除，直接删除 docker-compose rm -f # 删除服务容器挂载的数据卷 docker-compose rm -v # 删除工程中指定服务的容器 docker-compose rm -sv nginx 16.down 停止并删除所有服务的容器、网络、镜像、数据卷。 # 停止并删除工程中所有服务的容器、网络 docker-compose stop # 停止并删除工程中所有服务的容器、网络、镜像 docker-compose down --rmi all # 停止并删除工程中所有服务的容器、网络、数据卷 docker-compose down -v 17.images docker-compose images打印服务容器所对应的镜像。 # 打印所有服务的容器所对应的镜像 docker-compose images # 打印指定服务的容器所对应的镜像 docker-compose images nginx 18.port docker-compose port打印指定服务容器的某个端口所映射的宿主机端口。 docker-compose port nginx 80 19.top docker-compose top显示正在运行的进程。 # 显示工程中所有服务的容器正在运行的进程 docker-compose top # 显示工程中指定服务的容器正在运行的进程 docker-compose top nginx
标签：compose,target,bind,最全,source,docker,type From： https://www.cnblogs.com/Leonardo-li/p/16726622.html