• TDH-Search常用命令


    一、指令部分:

    1.search管理界面地址:
    2.集群状态查看命令:
    curl -XGET 'localhost:9200/_cluster/health?pretty'
    curl -XGET 'localhost:9200/_cat/health?v'
     
    检查shards unsigned原因
    curl 'localhost:9200/_cat/shards?h=index,shard,prirep,state,docs,store,ip,node,un*&v' | grep UNAS
    检查shards unsigned原因
    curl -XGET localhost:9200/_cat/shards?v | grep UNA
     
    3.查看集群的节点数目和主节点等信息,如下:
    curl -XGET 'localhost:9200/_cat/nodes?v'
    4.查看index的索引信息:
    curl -XGET 'localhost:9200/lxj.test_es_comm/_settings?pretty'
    curl -XGET 'localhost:9200/_clus ter/health/employee?pretty' (查看索引健康状态)
    curl -XGET 'localhost:9200/_cluster/health/?level=indices&pretty'(查看全部索引的健康状态)
    curl -XGET 'localhost:9200/_cluster/health/?level=shards&pretty' (查看分片级别的健康状态)
    curl -XPOST 'localhost:9200/lxj.test_es_comm/_open?pretty'(索引打开)
    curl -XPOST 'localhost:9200/lxj.test_es_comm/_close?pretty' (索引关闭)
    curl -s 'localhost:9200/_cat/indices?v' |grep red (查看red状态的索引)
     
    5.查看index的表结构:
    curl -XGET 'localhost:9200/default.test_people_es/_mappings?pretty'
    6.通过Curl使用RestAPI的格式:
    curl -X<VERB> '<HOST>:9200/<PATH>/[<API>]/[?<PARAMETERS>]' [-d '{<BODY>}']
    7.创建index:
    curl -XPUT '172.20.230.110:9200/test/?pretty'
    删除index:
    curl -XDELETE 'localhost:9200/test/?pretty'
    8.编入文档:(如果有会覆盖)
    curl -H "Content-Type: application/json" -XPUT 'localhost:9200/employee/dev/1?pretty' -d '{
    "firstname": "San",
    "lastname": "Zhang",
    "age": 26,
    "on_board_date": "2015-10-31",
    "hometown": "Beijing",
    "school": "Nanjing University",
    "married": false,
    "about": "I love Beijing Opera"
    }'
    输出:
    {
    "_index" : "employee", index名。
    "_type" : "dev", Type名。
    "_id" : "1", 这条Document的ID。。
    "_version" : 1, 版本号,每执行依次命令就加1。 
    "_shards" : { shard数目及状态。
    "total" : 2,
    "successful" : 2,
    "failed" : 0
    }, 
    "created" : true true表示第一次创建。 
    }
    9.查看/employee/dev/1下是否存在Document
    curl -i -XHEAD 'localhost:9200/employee/dev/1?pretty'
    10.获取/employee/dev/1下的Document
    curl -XGET 'localhost:9200/employee/dev/1?pretty'
    11.更新文档 
    curl -H "Content-Type: application/json"  -XPUT 'localhost:9200/employee/dev/1?pretty' -d '{
    "firstname": "aa",
    "lastname": "s",
    "age": 27,
    "on_board_date": "1995-02-02",
    "hometown": "南京",
    "school": "28所老年大学",
    "married": false,
    "about": "铁肩担大任,冲上山顶论英雄,联合起来办大事,做就做到最好,让创新成为习惯,共享才能共赢,创造幸福而有尊严的生活"
    }'
    12.获取/employee/dev/1中的 name 和 age 字段,多个字段用“,”隔开
    curl -XGET 'localhost:9200/employee/dev/1?_source=firstname,lastname,age&pretty'
    13.获取source部分的数据
    curl -XGET 'localhost:9200/employee/dev/1/_source?pretty'
     
    14.新建一个document,会自动生成rowkey:
    curl -H "Content-Type: application/json"  -XPOST 'localhost:9200/employee/sales?pretty' -d '{
    "firstname": "Lei",
    "lastname": "Li",
    "age": 28,
    "on_board_date": "2013-10-03",
    "hometown": "Hangzhou",
    "school": "Zhejiang University",
    "married": true,
    "about": "I appear in your English textbook."
    }'
    15.查询一条记录;
    curl -XGET 'localhost:9200/employee/_search?pretty&q=lastname:"wang"'
    16.设置副本数:
    curl -XPUT 'localhost:9200/employee/_settings?pretty' -d '{
    "number_of_replicas": 2
    }'
    16.设置刷新间隔(入库30秒后才能查询到数据,原本是入库一秒就可以查询到入库的数据,
    es要做很多合并的操作,会占用系统资源,降低入库速度)
    curl -XPUT 'localhost:9200/employee/_settings' -d '
    {
    "index" :{
    "refresh_interval" : "30s"
    }
    }
    '
    17.设置segment大小
    curl -XPOST 'localhost:9200/employee/_optimize?max_num_segments=3&pretty'
    18.设置批量插入的队列大小
    curl -XPUT 'localhost:9200/_cluster/settings' -d '
    {
    "persistent":{
    "indices.store.throttle.max_bytes_per_sec" : "100mb",
    "threadpool.bulk.queue_size" : "10000"
    }
    }
    '
    19.查看es集群nodes状态,是否挂掉
    curl -XGET 'localhost:9200/_cat/nodes?pretty'
    20.查看所有节点的统计数据
    curl -XGET 'localhost:9200/_nodes/stats?pretty'
    curl -XGET 'localhost:9200/_nodes/gz230-110/stats?pretty' (查看指定节点的状态)
    21.空检索
    curl -XGET 'localhost:9200/_search?pretty'
    22.指定条件检索
    curl -XGET 'localhost:9200/employee/sales/_search'
    curl -XGET 'localhost:9200/employee/sales,dev/_search'
    curl -XGET 'localhost:9200/employee/dev/_search?pretty&q=lastname:Li' (uri检索)
    23.请求体检索:
    curl -XPOST 'localhost:9200/employee/dev/_search?pretty' -d '
    {
    "query" : {
    "match_phrase" : {
    "lastname" : "Li"
    }
    }
    }
    '
    24.指定条件URI检索
    curl -XGET 'localhost:9200/employee/_search?q=Beijing&df=school'(默认查询字段为school)
     
    curl -XGET 'localhost:9200/employee/_search?q=school:(Beijing%20AND%20University)'(AND查询)
     
    curl -XGET 'localhost:9200/employee/_search?q=school:(Beijing%20OR%20University)'(OR查询)
     
    curl -XGET 'localhost:9200/employee/_search?q=*&from=1&size=10'(指定返回从第几条开始,几条结果)
     
    curl -XGET 'localhost:9200/default.test_people_es_1/_search?q=*&sort=id:asc&pretty' (排序)
     
    curl -XGET 'localhost:9200/employee/_search?q=(lastname:li)%20AND%20(school:beijing)'(查询条件AND)
    curl -XGET 'localhost:9100/employee/_search?q=+school:nanjing+school:beijing' (+操作符)
    curl -XGET 'localhost:9100/employee/_search?q=-school:university' (-操作符)
    curl -XGET 'localhost:9200/employee/_search?q=school=(beijing%20AND%20university)%20OR%20(na%20AND%20university)' (括号的使用)
    curl -XGET 'localhost:9200/employee/_search?q=??d*' (?匹配任意一个字符,*匹配任意个数的字符)
    curl -XGET 'localhost:9200/employee/_search?q=on_board_date:(>2014-01-01%20AND%20<2015-12-31)' (日期范围查询)
    25.按照querybody检索:
    curl -XGET '172.20.230.110:9200/default.test_people_es/_search?pretty' -d '{
    "query": {
    "bool" : {
    "must" : {
    "term":{
    "sex": "女"
    }
    }
    }
    }
    }'
     
    26.移动分片
    假设我们有两个节点:es_node_one和es_node_two,ElasticSearch在es_node_one节点上分配了ops索引的两个分片,
    我们现在希望将第二个分片移动到es_node_two节点上。可以如下操作实现:
    "commands" : [ {
    "move" : {
    "index" : "ops",
    "shard" : 1,
    "from_node" : "es_node_one",
    "to_node" : "es_node_two"
    }
    }]
    }'
     
    27.分配分片
    我们还可以将一个未分配的分片分配到一个指定的节点上。假设ops索引上有一个编号为0的分片尚未分配,
    并且我们希望ElasticSearch将其分配到es_node_two上,可以运行如下命令操作:
    "commands" : [ {
    "allocate" : {
    "index" : "ops",
    "shard" : 0,
    "node" : "es_node_two"
    }
    } ]
    }' 
     
    28.取消分配
    如果希望取消一个正在进行的分配过程,我们通过运行cancel命令来指定我们希望取消分配的索引、节点以及分片,如下所示:
    "commands" : [ {
    "cancel" : {
    "index" : "ops",
    "shard" : 0,
    "node" : "es_node_one"
    }
    } ]
    }' 
     
    创建索引:
    分配索引在某一节点:
    "index.routing.allocation.include.zone": "zone_one,zone_two"
    }'
     
    "transient" : {
    "cluster.routing.allocation.include._ip" "10.0.1.112,10.0.1.114"
    }
    }'
    设置每个节点上的分片为1:
    "index.routing.allocation.total_shards_per_node" : 1
    }'
     
    29.测试分词,analyzer=ik :用于指定分词器名;上述的四个分词器(english,standard,ik,mmseg)都可以指定
    curl -XPOST ‘localhost:9200/_analyze?analyzer=ik&&pretty’ -d '南京火车站' 
     
    30、新建索引时设定shards和replicas
    curl -XPUT http://localhost:9200/blog/ -d '{
    "settings" : {
    "number_of_shards" : 1,
    "number_of_replicas" : 2
    }
    }'

    31、内容修改
    curl -H "Content-Type: application/json" -XPOST '192.168.200.100:9200/employee/dev/1/_update?pretty' -d '{"hometown":"上海"}''   报错:
    "上海"}''
    {
      "error" : {
        "root_cause" : [
          {
            "type" : "action_request_validation_exception",
            "reason" : "Validation Failed: 1: script or doc is missing;"
          }
        ],
        "type" : "action_request_validation_exception",
        "reason" : "Validation Failed: 1: script or doc is missing;"
      },
      "status" : 400
    }

    解决:

     curl -H "Content-Type: application/json" -XPOST '192.168.200.100:9200/employee/dev/1/_update?pretty' -d '{"doc":{"hometown":"上海"}}'

     
    二、知识点说明部分:
    1.边框为深色的分片是index的主分片,其他为主分片的副本;
    2.方框中的数字表示同一索引的不同分片,相同数字表示同一分片的不同副本;
    3.索引的主分片的数目,建表后不可改;
    4.transwarp Search的含有三个特殊的数据对象:Index,Type,Document,Field,它们与传统二维表的映射关系如下:
    Index(索引)-->Table(表)
    Document-->Row(行)
    Field(字段)-->Column(列)
    Type 是Index的逻辑上的分类,不映射为传统二维表中的数据对象。
    5.如果一个Index的 Replica 数大于或等于集群中节点数量,这个Index中将会有分片无法分配到节点上;
    6.如果您需要在 <QUERY_STRING> 中包含以下Transwarp Search保留字符,您需要使用 进行转译:
    + - = && || > < ! ( ) { } [ ] ^ " ~ * ? : /
     
    三、注意事项:
    elasticsearch 6.x一个index下不支持创建多个type,并且官方说是在接下来的7.0版本中会删掉type
    如果在一个index下已经存在一个type会报错:
    [root@master ~]# curl -H "Content-Type: application/json" -XPOST 'localhost:9200/books/solr/1?pretty' -d '{"title":"Apache Solr 4 Cookbook",
    > "published": 2012}'
    {
      "error" : {
        "root_cause" : [
          {
            "type" : "illegal_argument_exception",
            "reason" : "Rejecting mapping update to [books] as the final mapping would have more than 1 type: [solr, es]"
          }
        ],
        "type" : "illegal_argument_exception",
        "reason" : "Rejecting mapping update to [books] as the final mapping would have more than 1 type: [solr, es]"
      },
      "status" : 400
    }

    四、一些参数

    1、关闭自动创建索引

    elasticsearch.yml配置文件中添加以下指令来关闭自动创建索引:
    action.auto_create_index: false

     

    curl -H "Content-Type: application/json" -XPOST '192.168.200.100:9200/employee/dev/1/_update?pretty' -d '{"doc":{"hometown":"上海"}}'

  • 相关阅读:
    学习进度条
    学习进度条
    《软件需求模式》读书笔记04
    大型网站技术架构阅读笔记5
    大型网站技术架构阅读笔记4
    大型网站技术架构阅读笔记3
    大型网站技术架构阅读笔记2
    大型网站技术架构阅读笔记1
    《uml大战需求分析》阅读笔记06
    《uml大战需求分析》阅读笔记05
  • 原文地址:https://www.cnblogs.com/yfb918/p/10416848.html
Copyright © 2020-2023  润新知