不积跬步无以至千里，不积小流无以至江海
不积跬步无以至千里，不积小流无以至江海

Docker-Compose部署Elk

Anonymous 收录于 Docker-Compose
 约 1000 字 预计阅读 5 分钟 

docker 安装ELK

1.下载项目 

git clone https://github.com/deviantony/docker-elk.git

2. 启动服务 

docker-compose up -d

3. 重置密码 

docker-compose exec -T elasticsearch bin/elasticsearch-setup-passwords auto --batch

输出如下：

Changed password for user apm_system
PASSWORD apm_system = m4TeQ4gEMJw3X5xaMv4B

Changed password for user kibana_system
PASSWORD kibana_system = RcBkkUCKRtNboXVm7NWO

Changed password for user kibana
PASSWORD kibana = RcBkkUCKRtNboXVm7NWO

Changed password for user logstash_system
PASSWORD logstash_system = WHbRnZhklNZf5apfOw2K

Changed password for user beats_system
PASSWORD beats_system = aRBr1Q8diQHAcHq8LGMn

Changed password for user remote_monitoring_user
PASSWORD remote_monitoring_user = XPfmZHbB7lnEsInKdSTC

Changed password for user elastic
PASSWORD elastic = DSLWlX5rOhFe5ta7Rd2i
4. 将密码妥善保存后，我们需要将 docker-compose.yml 配置文件中的 elasticsearch 服务的 ELASTIC_PASSWORD 去掉，这样可以确保服务启动只使用我们刚刚重置后的密码（keystore）。以及需要对 kibana 、 logstash 配置文件中的信息进行替换，将文件中的 elastic 用户的密码进行更新，相关文件我们在开篇的目录结构中有提过，暂时先修改下面三个文件就可以了：
  • kibana/config/kibana.yml
  • logstash/config/logstash.yml
  • logstash/pipeline/logstash.conf

新方式 

version: "3.7"
services: 
  # elk日志套件(镜像版本最好保持一致) ##########
  elasticsearch:
    container_name: elasticsearch
    image: elasticsearch:7.17.5
    restart: on-failure
    ports:
      - "9200:9200"
      - "9300:9300"
    environment:
      #单节点设置
      - discovery.type=single-node
      #锁住内存 提高性能
      - bootstrap.memory_lock=true 
      #设置启动内存大小 默认内存/最大内存
      - "ES_JAVA_OPTS=-Xms256m -Xmx256m" 
    ulimits:
      memlock:
        soft: -1
        hard: -1
    volumes: #挂载文件
      - ./elasticsearch/data:/usr/share/elasticsearch/data
      - ./elasticsearch/conf/elasticsearch.yml:/usr/share/elasticsearch/config/elasticsearch.yml
      - ./elasticsearch/logs:/usr/share/elasticsearch/logs
    networks:
      - elknetwork
  
  kibana:
    container_name: kibana
    image: kibana:7.17.5
    depends_on:
      - elasticsearch
    restart: on-failure
    ports:
      - "5601:5601"
    #volumes: 
      #- ./kibana/conf/kibana.yml:/usr/share/kibana/config/kibana.yml
    networks:
      - elknetwork
      
  logstash:
    container_name: logstash
    image: logstash:7.17.5
    depends_on:
      - elasticsearch
    restart: on-failure
    ports:
      - "9600:9600"
      - "5044:5044"
    volumes: #logstash.conf日志处理配置文件 
      - ./logstash/data/:/usr/share/logstash/data
      - ./logstash/conf/:/usr/share/logstash/conf/
      - ./logstash/pipeline:/usr/share/logstash/pipeline
    networks:
      - elknetwork
     
  filebeat:
    container_name: filebeat
    image: elastic/filebeat:7.17.5
    depends_on: 
      - elasticsearch
      - logstash
      - kibana
    restart: on-failure
    volumes:
      - ./filebeat/conf/filebeat.yml:/usr/share/filebeat/filebeat.yml
    networks:
      - elknetwork
#使用自定义的网桥名字
networks:
  elknetwork: 
    external: true

创建目录：

mkdir -p elasticsearch/{conf,data,logs,plugins} kibana/conf logstash/{conf,data,pipeline} filebeat/{conf,logs}
chmod 777 elasticsearch/{conf,data} logstash/{conf,data,pipeline}

./elasticsearch/conf/elasticsearch.yml

#主机ip
network.host: 0.0.0.0 
#network.bind_host: 0.0.0.0
#cluster.routing.allocation.disk.threshold_enabled: false
#node.name: es-master
#node.master: true
#node.data: true
#允许跨域，集群需要设置
http.cors.enabled: true 
#跨域设置
http.cors.allow-origin: "*" 
#http.port: 9200
#transport.tcp.port: 9300

./kibana/conf/kibana.yml

server.name: "kibana"
server.host: "0.0.0.0"
# 注意你的本地IP
elasticsearch.hosts: ["http://elasticsearch:9200"]
i18n.locale: "zh-CN"

./logstash/conf/logstash.yml

# 允许通过本机所有IP访问
http.host: "0.0.0.0"
# 指定使用管道ID
xpack.management.pipeline.id: ["main"]

./logstash/conf/jvm.options

-Xmx128m
-Xms128m

添加管道ID和管道配置文件目录映射，注意符号 - 前后都要有空格（巨坑）

./logstash/conf/pipelines.yml

 - pipeline.id: main
   path.config: "/usr/share/logstash/pipeline"

添 微服务应用日志管道配置，在上面指定了管道配置容器目录 /usr/share/logstash/pipeline , 后面启动 Logstash 时会将其挂载至宿主机目录 ./logstash/pipeline，接下来只要在宿主机目录下添加管道配置文件 youlai-log.config 就可以被 Logstash 自动加载生效。

./logstash/pipeline/elk-log.config

input {
 beats {
    port => 5044
    client_inactivity_timeout => 36000
  }
}
filter {
   mutate {
        remove_field => ["@version"]
        remove_field => ["tags"]
   }
}
output {
  if [appname] == "gateway" {
     elasticsearch {
       hosts => "http://elasticsearch:9200"
       index => "gateway-log"
     }
  }else if [appname] == "auth" {
     elasticsearch {
       hosts => "http://elasticsearch:9200"
       index => "auth-log"
     }
  }else if [appname] == "adm" {
     elasticsearch {
       hosts => "http://elasticsearch:9200"
       index => "adm-log"
     }
  }else if [appname] == "project" {
     elasticsearch {
       hosts => "http://elasticsearch:9200"
       index => "project-log"
     }
  }else if [appname] == "product" {
     elasticsearch {
       hosts => "http://elasticsearch:9200"
       index => "product-log"
     }
  }
  stdout {}
}

在日志收集服务器上面部署 filebeat

创建目录：

mkdir ./{conf,logs}
chmod 777 ./conf

docker-compose.yml

version: "3.7"
services: 
  filebeat:
    container_name: filebeat
    image: elastic/filebeat:7.17.5
    restart: on-failure
    volumes:
      - ./filebeat/conf/filebeat.yml:/usr/share/filebeat/filebeat.yml
      - /usr/local/springboot:/logs
    environment:
      --log-driver: json-file

注意：/usr/local/springboot 是运用服务器收集日志的文件夹

.filebeat/conf/filebeat.yml

filebeat.inputs:
- type: log
  enabled: true
  paths:
  - /logs/gateway/*.log
  fields:
    appname: gateway  # 自定义字段，提供给 Logstash 用于区分日志来源
  fields_under_root: true  # 开启自定义字段
- type: log
  enabled: true
  paths:
  - /logs/auth/*.log
  fields:
    appname: auth
  fields_under_root: true

processors:
  - drop_fields:
      fields: ["log","input","host","agent","ecs"] # 过滤不需要的字段

output.logstash:
	#公网ip
  hosts: ['139.155.xxx.xxx:5044']
filebeat.inputs:
- type: log
  enabled: true
  paths:
  - /logs/gateway/gateway.log
  fields:
    appname: gateway
  fields_under_root: true
- type: log
  enabled: true
  paths:
  - /logs/auth/auth.log
  fields:
    appname: auth
  fields_under_root: true

- type: log
  enabled: true
  paths:
  - /logs/adm/adm.log
  fields:
    appname: adm
  fields_under_root: true

- type: log
  enabled: true
  paths:
  - /logs/product/product.log
  fields:
    appname: product
  fields_under_root: true

processors:
  - drop_fields:
      fields: ["log","input","host","agent","ecs"]

output.logstash:
  hosts: ['139.155.xxx.xxx:5044']

服务器配置：

服务器 描述
139.155.xxx.xxx 部署ELK
47.94.xxx.xxx 部署FileBeat + 运用日志（日志收集 在哪台服务器，FileBeat 就部署在哪台服务器）

139.155.xxx.xxx 先创建目录：

mkdir -p /usr/local/docker/elk
cd /usr/local/docker/elk
mkdir -p elasticsearch/{conf,data,logs,plugins/ik} kibana/conf logstash/{conf,data,pipeline} 
# 需要给文件设置权限，否则启动报错（无权限）
chmod 777 elasticsearch/{conf,data} logstash/{conf,data,pipeline}

创建docker network

docker network create elknetwork

查看已创建的docker network

docker network ls

创建 docker-compose.yml 文件

version: "3.7"
services: 
  # elk日志套件(镜像版本最好保持一致)
  elasticsearch:
    container_name: elasticsearch
    image: elasticsearch:7.17.5
    restart: on-failure
    ports:
      - "9200:9200"
      - "9300:9300"
    environment:
      #单节点设置
      - discovery.type=single-node
      #锁住内存 提高性能
      - bootstrap.memory_lock=true 
      #设置启动内存大小 默认内存/最大内存
      - "ES_JAVA_OPTS=-Xms256m -Xmx256m" 
    ulimits:
      memlock:
        soft: -1
        hard: -1
    volumes:
      #挂载数据文件
      - ./elasticsearch/data:/usr/share/elasticsearch/data
      # 挂载配置文件
      - ./elasticsearch/conf/elasticsearch.yml:/usr/share/elasticsearch/config/elasticsearch.yml
      # 挂载IK中文分词器
      - ./elasticsearch/plugins/ik:/usr/share/elasticsearch/plugins/ik
    networks:
      - elknetwork
  
  kibana:
    container_name: kibana
    image: kibana:7.17.5
    depends_on:
      - elasticsearch
    restart: on-failure
    ports:
      - "5601:5601"
    volumes: 
      - ./kibana/conf/kibana.yml:/usr/share/kibana/config/kibana.yml
    networks:
      - elknetwork
      
  logstash:
    container_name: logstash
    image: logstash:7.17.5
    depends_on:
      - elasticsearch
    restart: on-failure
    ports:
      - "9600:9600"
      - "5044:5044"
    volumes:
      - ./logstash/data/:/usr/share/logstash/data
      - ./logstash/conf/:/usr/share/logstash/conf/
      - ./logstash/pipeline:/usr/share/logstash/pipeline
    networks:
      - elknetwork
     
#使用自定义的网桥名字
networks:
  elknetwork: 
    external: true
编写各个配置文件：

./elasticsearch/conf/elasticsearch.yml

#主机ip 允许所有IP访问
network.host: 0.0.0.0
node.name: "elasticsearch" 
##允许跨域，集群需要设置
http.cors.enabled: true 
##跨域设置
http.cors.allow-origin: "*"

./kibana/conf/kibana.yml

server.name: "kibana"
server.host: "0.0.0.0"
# 注意你的本地IP（我们在elasticsearch.yml中指定了 node.name: "elasticsearch" 也可以使用服务名：elasticsearch）
elasticsearch.hosts: ["http://elasticsearch:9200"]
# 汉化kibana（英语ok的忽略）
i18n.locale: "zh-CN"

./logstash/conf/logstash.yml

# 允许通过本机所有IP访问
http.host: "0.0.0.0"
# 指定使用管道ID
xpack.management.pipeline.id: ["main"]

./logstash/conf/jvm.options

-Xmx128m
-Xms128m

这个地方如果生产服务资源够多，也可以设置大一点或则使用默认的

添加管道ID和管道配置文件目录映射，注意符号 - 前后 都要有空格（巨坑）

./logstash/conf/pipelines.yml

 - pipeline.id: main
   path.config: "/usr/share/logstash/pipeline"

添加 微服务应用日志管道配置，在上面（pipelines.yml）指定了管道配置容器目录 /usr/share/logstash/pipeline , 后面启动 Logstash 时会将其挂载至宿主机目录 ./logstash/pipeline，接下来只要在宿主机目录下添加管道配置文件 elk-log.config 就可以被 Logstash 自动加载生效。

./logstash/pipeline/elk-log.config

注意该文件不是在 ./logstash/conf 目录下，在 ./logstash/pipeline 目录下

input {
 beats {
    port => 5044
    client_inactivity_timeout => 36000
  }
}
filter {
   mutate {
        remove_field => ["@version"]
        remove_field => ["tags"]
   }
}
output {
  if [appname] == "elk-gateway" {
     elasticsearch {
       hosts => "http://elasticsearch:9200"
       index => "elk-gateway-log"
     }
  }else if [appname] == "elk-auth" {
     elasticsearch {
       hosts => "http://elasticsearch:9200"
       index => "elk-auth-log"
     }
  }else if [appname] == "elk-adm" {
     elasticsearch {
       hosts => "http://elasticsearch:9200"
       index => "elk-adm-log"
     }
  }else if [appname] == "elk-project" {
     elasticsearch {
       hosts => "http://elasticsearch:9200"
       index => "elk-project-log"
     }
  }else if [appname] == "elk-product" {
     elasticsearch {
       hosts => "http://elasticsearch:9200"
       index => "elk-product-log"
     }
  }
  stdout {}
}

注意这里的 [appname] == “elk-gateway” 后面的名字是自定义的，但是在filebeat配置文件中需要一一对应起来，并且在fileBeat中添加了日志配置在这里也需要添加对应的标签才行。

到这里我们可以执行命令启动ELK服务了：

cd /usr/local/docker/elk
docker-compose up -d

如果拉取镜像的时间过长：可以终止掉使用docker pull命令先将镜像拉取在本地后再启动（我当时直接启动耗费了很长时间）

启动成功可已测试一下：

测试elasticsearch：

/images/docker-compose/elk-4.png

测试kibana

/images/docker-compose/elk-5.png

到此处：我们的ELK安装完毕

来到我们的运用服务器：47.94.xxx.xxx

我的日志在：/usr/local/springboot/ 目录下；运用是通过docket镜像的方式部署

/images/docker-compose/elk-6.png

创建目录：

mkdir -p /usr/local/docker/filebeat/{conf,logs}
chmod 777 /usr/local/docker/filebeat/conf

编写filebeat.yml文件

filebeat.inputs:
- type: log
  enabled: true
  paths:
  - /logs/elk-gateway/elk-gateway.log
  fields:
    appname: elk-gateway
  fields_under_root: true
- type: log
  enabled: true
  paths:
  - /logs/elk-auth/elk-auth.log
  fields:
    appname: elk-auth
  fields_under_root: true

- type: log
  enabled: true
  paths:
  - /logs/elk-adm/elk-adm.log
  fields:
    appname: elk-adm
  fields_under_root: true

- type: log
  enabled: true
  paths:
  - /logs/elk-project/elk-project.log
  fields:
    appname: elk-project
  fields_under_root: true


- type: log
  enabled: true
  paths:
  - /logs/elk-product/elk-product.log
  fields:
    appname: elk-product
  fields_under_root: true

processors:
  - drop_fields:
      fields: ["log","input","host","agent","ecs"]
# 将日志输出到 logstash
output.logstash:
  hosts: ['139.155.xxx.xxx:5044']

注意这里的 paths: /logs/xxx/xxx.log 是容器内部的路径不是宿主机的日志目录

编写docker-compose.yml文件

version: "3.7"
services: 
  filebeat:
    container_name: filebeat
    image: elastic/filebeat:7.17.5
    restart: on-failure
    volumes:
      - ./conf/filebeat.yml:/usr/share/filebeat/filebeat.yml
      - /usr/local/springboot:/logs
    environment:
      --log-driver: json-file

==这里将宿主机的日志目录 /usr/local/springboot 挂载到容器中去，这里很关键，如果配置错误，在kibana中看不见索引。==

然后启动镜像

配置kibana

登录：http://139.155.xxx.xxx:5601/

后面根据图中步骤配置即可

/images/docker-compose/elk-7.png

/images/docker-compose/elk-8.png

/images/docker-compose/elk-9.png

/images/docker-compose/elk-11.png

/images/docker-compose/elk-12.png

更新于 2023-08-12 
Docker-ComposeElk
返回 | 主页
0%