跳至主要內容

elasticsearch

OrangBus大约 4 分钟

当使用httpclient 和 elasticsearch 的时候，他们两个的版本需要保持一致

index：database

type: table

doc: 一行记录

版本选择：

跨域处理

# vim /elasticsearch/config/elasticsearch.yml

http.cors.enabled: true
http.cors.allow-origin: "*"

# https 访问关闭
xpack.security.enabled: false

单节点docker安装

插件安装

./bin/elasticsearch-plugin install https://github.91chi.fun/https://github.com//medcl/elasticsearch-analysis-ik/releases/download/v8.1.0/elasticsearch-analysis-ik-8.1.0.zip

mysql-jdbc

https://mvnrepository.com/artifact/mysql/mysql-connector-java/8.0.17

cerebro

集群安装

master

cluster.name: orangbus
node.name: master
node.master: true
network.host: 0.0.0.0 # 指定ip

node1

cluster.name: orangbus # 主节点保持一致
http.port: 8200
discovery.zen.ping.unicast.host: ["127.0.0.1"]

node2

cluster.name: orangbus # 主节点保持一致
http.port: 8300
discovery.zen.ping.unicast.host: ["127.0.0.1"]

mysql配置文件

input {
  jdbc{
    jdbc_driver_library => "D:\\laragon\\www\\Java\\elk\\mysql-connector-java-8.0.16.jar"
    jdbc_driver_class => "com.mysql.cj.jdbc.Driver"
    jdbc_connection_string => "jdbc:mysql://127.0.0.1:3306/laravel_web?characterEncoding=utf8&useSSL=false&serverTimezone=UTC&rewriteBatchedStatements=true"
    jdbc_validate_connection => true
    jdbc_user => "root"
    jdbc_password => "root"
    schedule => "* * * * *"
    clean_run => true
    statement => "SELECT * FROM jokes WHERE updated_at >= :sql_last_value"
    jdbc_paging_enabled => "true"
    jdbc_page_size => "1000"  # 每页1000条数据
    jdbc_default_timezone =>"Asia/Shanghai"
  }
}

output {
  elasticsearch {
    hosts => ["http://162.14.72.65:9200"]
    index => "jokes"
    document_id => "%{id}"
  }
}

创建配置文件logstash_mysql.conf

input{
     jdbc {  # java数据库访问的API接口
         jdbc_driver_library => "/home/python/mysql-connector-java-8.0.13/mysql-connector-java-8.0.13.jar"
         jdbc_driver_class => "com.mysql.jdbc.Driver"
         jdbc_connection_string => "jdbc:mysql://127.0.0.1:3306/toutiao?tinyInt1isBit=false"
         jdbc_user => "root"
         jdbc_password => "mysql"
         jdbc_paging_enabled => "true"  # 数据分页, 一共14W数据
         jdbc_page_size => "1000"  # 每页1000条数据
         jdbc_default_timezone =>"Asia/Shanghai"
         statement => "select a.article_id as article_id,a.user_id as user_id, a.title as title, a.status as status, a.create_time as create_time,  b.content as content from news_article_basic as a inner join news_article_content as b on a.article_id=b.article_id"  # 联表查询, 尽量起别名,否则ES的字段名称会变为a.xx, 这样和mysql的字段名称会出现差异
     }
}

output{
      elasticsearch {
         hosts => "127.0.0.1:9200"
         index => "articles"
         document_id => "%{article_id}"  # 让文档id记录文章id, 方便进行数据库查询
         document_type => "article"
      }
      stdout {  # 导入过程中以json形式显式的输出导入的内容
         codec => json_lines  
     }
}
  • 增量更新的配置
input{
  record_last_run => "true"  	# 记录最后一次运行时的数据点, 默认为最后一次更新的时间
	use_column_value => "true"  # 不再记录最后一次更新的时间, 而是记录最后一次更新时, 数据库某个字段的值(字段的值要求是递增的)
  tracking_column => "article_id"  # 设置记录的字段
last_run_metadata_path => "/xx/data"    # 数据点的存储位置
  clean_run => "false"    # 从存储位置开始继续读取, 如果设置为true, 则清除数据点, 从头开始读取
}

错误汇总

数据库链接失败

把mysql-connector-java-8.0.16.jar复制到logstash-7.1.0\logstash-core\lib\jars

驱动加载失败

基本操作

methodapi案例描述
PUT/index/_doc/id/joke/_doc/1创建索引（指定id）
POST/index/_doc/joke/_doc创建索引 （随机id）
POST/index/_doc/_update/joke/_doc/_update修改文档
DELETE/index/_doc/id/joke/_doc/1删除文档
GET/index/_doc/id/joke/_doc/1通过id查询文档
POST/index/_doc/_search/joke/_doc/_search搜索文档

索引（数据库）

创建索引

PUT /indexName

编辑

删除

DELETE /indexname

分词测试

GET /jokes/_analyze
{
  "analyzer": "standard",
  "text": "搞笑"
}

查询索引状态

GET /jokes/_stats

文档操作（表）

添加文档

PUT /index/_doc/id

PUT /videos/_doc/id
{
	"title":"mysql从select到delete",
	"url":"http://doc.orangbus.cn"
}

编辑文档

删除文档

docker镜像加速

sudo mkdir -p /etc/docker
sudo tee /etc/docker/daemon.json <<-'EOF'
{
  "registry-mirrors": ["https://0czkbjwu.mirror.aliyuncs.com"]
}
EOF
sudo systemctl daemon-reload
sudo systemctl restart docker

logstash -f mysql.conf

查询

text: 会使用分词器解析

keyword: 不会使用分词器解析

GET _analyze
{
  "analyzer": "standard",
  "text": "老司机"
}

GET _analyze
{
  "analyzer": "keyword",
  "text": "老司机"
}

全部匹配

match: 会使用分词器

profile: true 显示查询语句

opearator: and 同时包含查询

match_phrase: 有顺序要求

term : 不对关键词进行分词处理

GET /jokes/_search
{
  "query": {
    "match": {
      "title" :"搞笑"
    }
  }
}

分页查询

GET /jokes/_search
{
  "query": {
    "match": {
      "title" :"搞笑"
    }
  },
  "form":0,
  "size":10
}

bool查询

msut ( and )

shoud ( or )

GET /jokes/_search
{
  "query": {
    "bool": {
      "must" :[
      	{
      		"match":{
      		"title":"搞笑"
      		}
      	}
      ]
    }
  }
}

查询 id<10

GET /jokes/_search
{
  "query": {
    "bool": {
      "must" :[
      	{
      		"match":{
      		"title":"搞笑"
      		}
        }
      ],
      "filter": [
        {
          "range": {
            "id": {
              "lte": 10
            }
          }
        }
      ]
    }
  }
}

多条件查询

空格分页

GET /jokes/_search
{
  "query": {
    "match": {
      "author": "笑话 搞笑"
    }
  },
  "_source": ["title","author"]
}

精确查询-term

GET /jokes/_search
{
  "query": {
    "term": {
      "title": "笑话大全 爆笑短笑话,听话"
    }
  },
  "_source": ["title","author"]
}

排序:没有分值

GET /jokes/_search
{
    "from":0,
    "size":100,
    "query":{
        "match":{
            "title":"搞笑"
        }
    },
    "_source": ["title","author","form_url","created_at","updated_at"],
    "sort": [
      {
        "updated_at": {
          "order": "desc"
        }
      }
    ]
}

字段限制

GET /jokes/_search
{
  "query": {
    "match": {
      "title" :"搞笑"
    }
  },
  "_source": ["title","author"]
}

高亮查询

GET /test/_search
{
  "query": {
    "match": {
      "name": "orangbus"
    }
  },
  "highlight": {
    "pre_tags": "<span style='color:red'>", 
    "post_tags": "</p>", 
    "fields":{
      "name": {}
    }
  }
}

ik分词器

ik_smart: 最少切分

ik_max_word: 最细颗粒

elasticsearch集成

集成的版本要和当前连接的 elasticsearch的版本保持一致

错误汇总

OpenJDK 64-Bit Server VM warning: INFO: os::commit_memory(0x00000005c2800000, 8547991552, 0) failed; error='???????С?????????????' (DOS error/errno=1455)

内存不足：

修改elasticsearch-8.1.0\config\jvm.options

-Xms1g
-Xmx1g

Elasticsearch无法访问

修改elasticsearch-8.1.0\config\elasticsearch.yaml

http.port: 9200
network.host: 0.0.0.0

设计需求

超级管理员

普通管理员

调拨人员

橙子工作室
Copyright © 2024 OrangBus