环境:
OS:Centos 7
ES(源端和目标端):6.8.5
logstash:6.8.5
说明:
1.logstash版本尽量与es版本保持一致
2.我们这里先在目的端创建好mapping(表结构),因为不提前创建好mapping的话,logstash同步的时候根据数据情况自动推断数据字段类型定义字段类型,
创建的mapping与原来的可能不一致,比如将源端的short类型同步后变成了long类型,个别情况会把text类型变成long类型，导致最后的同步会失败.
1.查看源端的index
[elasticsearch@localhost bin]$ curl -u elastic:elastic -X GET "192.168.1.108:19200/_cat/indices?v" health status index uuid pri rep docs.count docs.deleted store.size pri.store.size green open .security-6 TBWap6LdRY6ALA4vDIagew 1 0 6 0 19.5kb 19.5kb yellow open app_message_all C04l_PP1RQyFPNBSf1cjjg 5 1 1000000 0 766.2mb 766.2mb
2.获取源端index的mapping
[elasticsearch@localhost bin]$ curl -u elastic:elastic -H "Content-Type: application/json" -XGET "http://192.168.1.108:19200/app_message_all/_mappings?pretty=true"
{
"app_message_all" : {
"mappings" : {
"_doc" : {
"properties" : {
"create_time" : {
"type" : "text",
"fields" : {
"keyword" : {
"type" : "keyword",
"ignore_above" : 256
}
}
},
"deleted" : {
"type" : "long"
},
"extra" : {
"type" : "text",
"fields" : {
"keyword" : {
"type" : "keyword",
"ignore_above" : 256
}
}
},
"massive_type" : {
"type" : "long"
},
"message" : {
"type" : "text",
"fields" : {
"keyword" : {
"type" : "keyword",
"ignore_above" : 256
}
}
},
"message_id" : {
"type" : "long"
},
"message_type" : {
"type" : "long"
},
"send_date" : {
"type" : "text",
"fields" : {
"keyword" : {
"type" : "keyword",
"ignore_above" : 256
}
}
},
"sender_seq_no" : {
"type" : "text",
"fields" : {
"keyword" : {
"type" : "keyword",
"ignore_above" : 256
}
}
},
"status" : {
"type" : "long"
},
"title" : {
"type" : "text",
"fields" : {
"keyword" : {
"type" : "keyword",
"ignore_above" : 256
}
}
},
"user_id" : {
"type" : "long"
}
}
}
}
}
}
这里index名称叫:app_message_all
type叫:_doc
说明:7.0之前是可以自定义type名称的,7.0之后的版本不能自定义type名称了,全部统一是_doc
3.目标端创建index和mapping
首先创建一个空的index(没有任何字段信息)
[elasticsearch@localhost bin]$curl -u elastic:elastic123 -X PUT "192.168.1.109:19200/app_message_all?pretty" -H 'Content-Type: application/json' -d'
{}
'
查询这个时候是没有任何的mapping信息的
[elasticsearch@localhost bin]$curl -u elastic:elastic123 -H "Content-Type: application/json" -XGET "http://192.168.1.109:19200/app_message_all/_mappings?pretty=true"
添加mapping信息,这里只需要将源端的properties对应的各项拷贝过来即可,注意{}个数要一致，结尾的4个}去掉
这里需要指定type=_doc需要与源端一致,6版本的type是可以自定义的
curl -u elastic:elastic123 -H 'Content-Type: application/json' -XPOST "http://192.168.1.109:19200/app_message_all/_doc/_mapping?pretty" -d '
{
"properties" : {
"create_time" : {
"type" : "text",
"fields" : {
"keyword" : {
"type" : "keyword",
"ignore_above" : 256
}
}
},
"deleted" : {
"type" : "long"
},
"extra" : {
"type" : "text",
"fields" : {
"keyword" : {
"type" : "keyword",
"ignore_above" : 256
}
}
},
"massive_type" : {
"type" : "long"
},
"message" : {
"type" : "text",
"fields" : {
"keyword" : {
"type" : "keyword",
"ignore_above" : 256
}
}
},
"message_id" : {
"type" : "long"
},
"message_type" : {
"type" : "long"
},
"send_date" : {
"type" : "text",
"fields" : {
"keyword" : {
"type" : "keyword",
"ignore_above" : 256
}
}
},
"sender_seq_no" : {
"type" : "text",
"fields" : {
"keyword" : {
"type" : "keyword",
"ignore_above" : 256
}
}
},
"status" : {
"type" : "long"
},
"title" : {
"type" : "text",
"fields" : {
"keyword" : {
"type" : "keyword",
"ignore_above" : 256
}
}
},
"user_id" : {
"type" : "long"
}
}
}'
这个时候可以比较下源端和目的端的mapping是否一致
源端
curl -u elastic:elastic -H "Content-Type: application/json" -XGET "http://192.168.1.108:19200/app_message_all/_mappings?pretty=true"
目的端
curl -u elastic:elastic123 -H "Content-Type: application/json" -XGET "http://192.168.1.109:19200/app_message_all/_mappings?pretty=true"
4.logstach配置文件
[root@localhost config]# more sync_all_index.conf
input {
elasticsearch {
hosts => ["http://192.168.1.108:19200"]
index => "*,-.monitoring*,-.security*,-.kibana*" ##我这里去掉了系统的index,目前我这里只有app_message_all,所以使用了*,也可以具体指定index名称的
user => "elastic"
password => "elastic"
size => 1000
scroll => "1m"
docinfo => true
}
}
# 该部分被注释，表示filter是可选的
filter {
mutate {
remove_field => ["@timestamp", "@version"] #过滤掉logstash 自己加上的字段
}
}
output {
elasticsearch {
hosts => ["http://192.168.1.109:19200"]
user => "elastic"
password => "elastic123"
index => "%{[@metadata][_index]}"
document_type => "%{[@metadata][_type]}"
document_id => "%{[@metadata][_id]}"
}
}
5.同步
[root@localhost config]# /opt/logstash-6.8.5/bin/logstash -f /opt/logstash-6.8.5/config/sync_all_index.conf
验证:
1.发现使用logstash同步过来的数据字段顺序与源端是不一致的
_id是eVpxaYsBeUT5Hwt-LE-o 查出来的数据字段顺序不一致
[elasticsearch@localhost bin]$ curl -u elastic:elastic -XGET 'http://192.168.1.108:19200/app_message_all/_doc/eVpxaYsBeUT5Hwt-LE-o?pretty'
[elasticsearch@localhost bin]$ curl -u elastic:elastic123 -XGET 'http://192.168.1.109:19200/app_message_all/_doc/eVpxaYsBeUT5Hwt-LE-o?pretty'
2.目的端创建maping的时候可以少一些字段,但是同步会自动添加少的那些字段,若想不同步该字段可以通过filter实现
filter {
mutate {
remove_field => ["@timestamp", "@version","user_id"]
}
}
标签：elastic,keyword,端先,text,mapping,long,message,type,logstash From： https://www.cnblogs.com/hxlasky/p/17788936.html