如何在docker基础上使用sh脚本实现项目部署

原创已于 2022-03-07 18:02:18 修改 · 4.7k 阅读

8 ·

CC 4.0 BY-SA版权

文章标签：

#运维开发

于 2022-03-07 17:54:07 首次发布

后端专栏收录该内容

146 篇文章

订阅专栏

本文详细介绍了如何利用Docker和Docker-Compose在Linux环境下部署微服务项目。通过docker-compose脚本（如app.yml、service.yml、env.yml）配置，实现不同微服务的启动，包括对外暴露端口、挂载数据卷、设置环境变量等。同时，展示了Dockerfile用于构建镜像，以及.sh脚本（如xk-api.sh）用于启动和更新微服务。此外，还提到了如何通过xk-publish.sh脚本批量启动微服务，整个流程旨在确保服务快速恢复或新版本发布。

前提须知:要安装docker和docker-compose才行,具体参考可跳转至如何安装和使用docker和如何安装和使用docker-compose

Linux目录架构演示

以下每个文件夹对应一个微服务项目,以api文件夹为例演示
以下.yml文件为docker-compose的对应配置文件, 主要目的是在不需要重新编译jar包,不需要重新打包jar的情况下重启服务而用(不需要更新新版本情况下,机子崩了,项目都停了,要重新启动项目)
以下.sh文件是脚本部署文件,主要用于代码更新后新打包的jar包去发布部署上线(代码更新了新功能,发布新版本,要置换掉原来jar包后才使用该脚本)
演示api微服务项目的对应的文件夹

docker-compose脚本

以下的app.yml和service.yml性质是一样的,只是因为演示的项目构造的原因分成2个文件而已(参考其中一个就行),用于启动微服务项目
以下的env.yml用于启动一些中间件配置服务的启动(比如nginx那些)
让现有的镜像变成容器跑起来

docker-compose-app.yml

version: '3'
services:
  xk-auth:
    image: xk/xk-auth:1.0-SNAPSHOT
    container_name: xk-auth
    ports:
      - 6689:6689
    volumes:
      - /data/logs/xk-auth:/logs/xk-practice-auth
    environment:
      - 'TZ="Asia/Shanghai"'
    external_links:
      - nacos-registry:nacos-registry #可以用nacos-registry这个域名访问nacos服务
  xk-act:
    image: xk/xk-act:1.0-SNAPSHOT
    container_name: xk-act
    ports:
      - 6687:6687
    volumes:
      - /data/logs/xk-act:/logs/xk-practice-act
    environment:
      - 'TZ="Asia/Shanghai"'
    external_links:
      - nacos-registry:nacos-registry #可以用nacos-registry这个域名访问nacos服务
#      - redis:redis
      - elasticsearch:es
  xk-ach:
    image: xk/xk-ach:1.0-SNAPSHOT
    container_name: xk-ach
    ports:
      - 6683:6683
    volumes:
      - /data/logs/xk-ach:/logs/xk-practice-ach
    environment:
      - 'TZ="Asia/Shanghai"'
    external_links:
      - nacos-registry:nacos-registry #可以用nacos-registry这个域名访问nacos服务
#      - redis:redis
      - elasticsearch:es
  xk-oact:
    image: xk/xk-oact:1.0-SNAPSHOT
    container_name: xk-oact
    ports:
      - 6682:6682
    volumes:
      - /data/logs/xk-oact:/logs/xk-practice-oact
    environment:
      - 'TZ="Asia/Shanghai"'
    external_links:
      - nacos-registry:nacos-registry #可以用nacos-registry这个域名访问nacos服务
#      - redis:redis
      - elasticsearch:es
  xk-rpt:
    image: xk/xk-rpt:1.0-SNAPSHOT
    container_name: xk-rpt
    ports:
      - 6688:6688
    volumes:
      - /data/logs/xk-rpt:/logs/xk-practice-rpt
    environment:
      - 'TZ="Asia/Shanghai"'
    external_links:
      - nacos-registry:nacos-registry #可以用nacos-registry这个域名访问nacos服务
#      - redis:redis
      - elasticsearch:es
  xk-search:
    image: xk/xk-search:1.0-SNAPSHOT
    container_name: xk-search
    ports:
      - 6686:6686
    volumes:
      - /data/logs/xk-search:/logs/xk-practice-search
    environment:
      - 'TZ="Asia/Shanghai"'
    external_links:
      - nacos-registry:nacos-registry #可以用nacos-registry这个域名访问nacos服务
#      - redis:redis
      - elasticsearch:es
  xk-sys:
    image: xk/xk-sys:1.0-SNAPSHOT
    container_name: xk-sys
    ports:
      - 6684:6684
    volumes:
      - /data/logs/xk-sys:/logs/xk-practice-sys
    environment:
      - 'TZ="Asia/Shanghai"'
    external_links:
      - nacos-registry:nacos-registry #可以用nacos-registry这个域名访问nacos服务
#      - redis:redis
      - elasticsearch:es
  xk-canal:
    image: xk/xk-canal:1.0-SNAPSHOT
    container_name: xk-canal
    ports:
      - 7788:7788
    volumes:
      - /data/logs/xk-canal:/logs/xk-practice-canal-customer
    environment:
      - 'TZ="Asia/Shanghai"'
    external_links:
      - nacos-registry:nacos-registry #可以用nacos-registry这个域名访问nacos服务
#      - redis:redis
      - elasticsearch:es

docker-compose-service.yml

version: '3'
services:
  xk-admin:
    image: xk/xk-admin:1.0-SNAPSHOT
    container_name: xk-admin
    ports:
      - 7616:7616
    volumes:
      - /data/logs/xk-admin:/logs/xk-practice-admin-application
      - /data/resource:/data/resource
    environment:
      - 'TZ="Asia/Shanghai"'
    external_links:
      - nacos-registry:nacos-registry #可以用nacos-registry这个域名访问nacos服务
#      - redis:redis
      - elasticsearch:es
  xk-portal:
    image: xk/xk-portal:1.0-SNAPSHOT
    container_name: xk-portal
    ports:
      - 7618:7618
    volumes:
      - /data/logs/xk-portal:/logs/xk-practice-portal-application
      - /data/xk/portal:/data/xk/portal
    environment:
      - 'TZ="Asia/Shanghai"'
    external_links:
      - nacos-registry:nacos-registry #可以用nacos-registry这个域名访问nacos服务
#      - redis:redis
      - elasticsearch:es
  xk-api:
    image: xk/xk-api:1.0-SNAPSHOT
    container_name: xk-api
    ports:
      - 7778:7778
    volumes:
      - /data/logs/xk-api:/logs/xk-practice-api
    environment:
      - 'TZ="Asia/Shanghai"'
    external_links:
      - nacos-registry:nacos-registry #可以用nacos-registry这个域名访问nacos服务
#      - redis:redis
      - elasticsearch:es
  xk-gateway:
    image: xk/xk-gateway:1.0-SNAPSHOT
    container_name: xk-gateway
    ports:
      - 7777:7777
    volumes:
      - /data/logs/xk-gateway:/logs/xk-practice-gateway
    environment:
      - 'TZ="Asia/Shanghai"'
    external_links:
      - nacos-registry:nacos-registry #可以用nacos-registry这个域名访问nacos服务

docker-compose-env.yml

version: '3'
services:
  nginx:
    image: nginx
    container_name: nginx
    volumes:
      - /data/nginx/conf:/etc/nginx #配置文件挂载
      - /data/nginx/html:/usr/share/nginx/html #静态资源根目录挂载
      - /data/nginx/log:/var/log/nginx #日志文件挂载
    ports:
      - 80:80
      - 8889:8889
  elasticsearch:
    image: elasticsearch:6.8.0
    container_name: elasticsearch
    environment:
      - "cluster.name=elasticsearch" #设置集群名称为elasticsearch
      - "discovery.type=single-node" #以单一节点模式启动
      - "ES_JAVA_OPTS=-Xms512m -Xmx512m" #设置使用jvm内存大小
    volumes:
      - /data/elasticsearch/plugins:/usr/share/elasticsearch/plugins #插件文件挂载
      - /data/elasticsearch/data:/usr/share/elasticsearch/data #数据文件挂载
    ports:
      - 9200:9200
  kibana:
    image: kibana:6.8.0
    container_name: kibana
    links:
      - elasticsearch:es #可以用es这个域名访问elasticsearch服务
    depends_on:
      - elasticsearch #kibana在elasticsearch启动之后再启动
    environment:
      - "elasticsearch.hosts=http://es:9200" #设置访问elasticsearch的地址
    ports:
      - 5601:5601
  logstash:
    image: logstash:6.8.0
    container_name: logstash
    volumes:
      - /data/log-stash/logstash-springboot.conf:/usr/share/logstash/pipeline/logstash.conf 
    depends_on:
      - elasticsearch #kibana在elasticsearch启动之后再启动
    links:
      - elasticsearch:es #可以用es这个域名访问elasticsearch服务
    ports:
      - 4560:4560
  nacos-registry:
    image: nacos/nacos-server:1.3.0
    container_name: nacos-registry
    volumes:
      - /data/nacos-registry:/home/nacos/logs
    environment:
      - "MODE=standalone"
    env_file:
      - /data/nacos-registry/env/nacos-standlone-mysql.env
    ports:
      - 8848:8848
#  redis:
#    image: redis:5.0.8
#    container_name: redis
#    command: redis-server --requirepass rediss --appendonly yes
#    volumes:
#      - /data/redis/data:/data #数据文件挂载
#      - /data/redis/etc/redis.conf:/usr/local/etc/redis/redis.conf
#      - /data/redis/logs:/logs
#    ports:
#      - 6379:6379

启动具体某个微服务

在docker-compose的yml配置文件的当前目录下输入以下命令即可指定启动配置文件中某个项目
这里是启动docker-compose-service.yml配置文件里配置的api微服务

docker-compose -f docker-compose-service.yml up -d xk-api

具体微服务jar包对应目录的配置脚本(重点)

以下演示微服务为api
将自己微服务项目的jar包,放在当前目录下即可

Dockerfile

FROM java:openjdk-8-jdk-alpine
ADD *.jar /home/api.jar

ARG JAVA_OPTS="-Xmx1344M -Xms1344M -Xmn448M -XX:MaxMetaspaceSize=192M -XX:MetaspaceSize=192M"
ARG JVM_SERVER_FIXED_ARGS="-server \
-XX:+DisableExplicitGC \
-XX:+UseParNewGC \
-XX:+UseConcMarkSweepGC \
-XX:+UseCMSInitiatingOccupancyOnly \
-XX:CMSInitiatingOccupancyFraction=70 \
-XX:+ExplicitGCInvokesConcurrentAndUnloadsClasses \
-XX:+CMSClassUnloadingEnabled \
-XX:+ParallelRefProcEnabled \
-XX:+CMSScavengeBeforeRemark \
-XX:+HeapDumpOnOutOfMemoryError \
-XX:+PrintGCDetails \
-XX:+PrintGCTimeStamps \
-XX:+PrintHeapAtGC \
-XX:+PrintGCApplicationStoppedTime \
-XX:+PrintGCDateStamps \
-XX:ErrorFile=/var/log/hs_err_pid%p.log \
-XX:HeapDumpPath=/var/log -Xloggc:/var/log/gc%t.log"
#ARG JAVA_AGENT="-javaagent:/usr/skywalking/agent/skywalking-agent.jar -Dskywalking.agent.service_name=bonade-hrm-gateway -Dskywalking.plugin.mysql.trace_sql_parameters=true -Dskywalking.collector.backend_service=127.0.0.1:11800"
ENV TZ=Asia/Shanghai
ENV JAVA_OPTS=${JAVA_OPTS}
ENV JAVA_AGENT=${JAVA_AGENT}
ENV JVM_SERVER_FIXED_ARGS=${JVM_SERVER_FIXED_ARGS}
ENTRYPOINT exec java ${JAVA_OPTS} ${JVM_SERVER_FIXED_ARGS} ${JAVA_AGENT} -jar /home/api.jar

xk-api.sh(启动微服务项目,常用)

输入命令到当前目录下, 直接输入./xk-api.sh 即可直接将当前目录下的jar包变成docker方式启动
使jar包变成镜像,然后将镜像变成容器跑起来


#!/usr/bin/env bash
app_name='xk-api'
docker build -t xk/${app_name}:1.0-SNAPSHOT .
docker stop ${app_name}
echo '----stop container----'
docker rm ${app_name}
echo '----rm container----'
docker rmi `docker images | grep none | awk '{print $3}'`
echo '----rm none images----'
docker run -p 7778:7778 --name ${app_name} \
--link nacos-registry:nacos-registry \
--link elasticsearch:es \
--link logstash:logstash \
--network xk_default \
-e TZ="Asia/Shanghai" \
-v /etc/localtime:/etc/localtime \
-v /data/logs/${app_name}:/logs/xk-practice-api \
-v /data/act_file/inventory:/data/act_file/inventory \
-v /data/act_file/plan:/data/act_file/plan \
-d xk/${app_name}:1.0-SNAPSHOT
echo '----start container----'

xk-publish.sh(更新启动所有微服务)

这里演示的是启动所有微服务项目跑起来,除了admin和portal的微服务项目


#!/usr/bin/env sh
for file in ./*
do
  if [ -d $file ]
   then
     if [ $file == './admin' ] ||[ $file == './portal' ]
      then
         echo $file '当前服务需要在auth 服务启动后再启动，请手动执行'
      else
        cd $file
        sh *.sh
        cd ../
     fi
  fi
done