티스토리 뷰
[업데이트 2017.02.14 11:15]
Elasticsearch로 입력하는 데이터의 field가 많고, 특정 field는 number로 변환이 되어야 하는 경우에 _default_ mapping에 numeric_detection를 true로 설정하여 사용합니다. 그러나 version 체계(예: 1.00)를 명시하는 field의 경우 Elasticsearch에 입력시 잘못된 데이터 타입 변환으로 아래와 같이 오류가 발생할 때가 있습니다.
:response=>
{
"index"=>
{"_index"=>"logstash-data-2016.01",
"_type"=>"data",
"_id"=>"14173638",
"status"=>400,
"error"=>
{ "type"=>"illegal_argument_exception",
"reason"=>"mapper [Data.OSVersion] of different type, current_type [text], merged_type [float]"
}
}
}
}
그럴때는 추가로 dynamic_templates를 이용하여 명시적으로 text type로 변환되도록 지정합니다. path_match를 이용하여 해당 데이터 field에 대해 인식 및 변환되도록 지정하였습니다.
PUT _template/logstash_data
{
"template": "logstash-data-*",
"settings": {
"index": {
"number_of_shards": "3",
"number_of_replicas": "1",
"mapping": {
"total_fields": {
"limit": "4000"
}
}
}
},
"mappings": {
"_default_": {
"numeric_detection": true,
"dynamic_templates": [
{
"texts": {
"path_match": "Data.OSVersion",
"mapping": {
"type": "text"
}
}
}
],
"_all": {
"enabled": true
}
}
}
}
Elastichsearch에 데이터 입력시 numeric detection > dynamic_templates 순으로 데이터 맵핑이 적용 됨을 확인 할 수 있습니다.
댓글
공지사항
최근에 올라온 글
최근에 달린 댓글
- Total
- Today
- Yesterday
링크
TAG
- Badge
- belief
- #TensorFlow
- #ApacheSpark
- Library
- English
- ate
- Jekyll and Hyde
- sentence test
- Physical Simulation
- #ELK Stack
- Meow
- 2D Game
- SSM
- project
- #ApacheZeppelin
- docker
- Mask R-CNN
- ILoop Engine
- Ragdoll
- Worry
- 도커
- Sea Bottom
- Memorize
- some time ago
- #REST API
- GOD
- OST
- aws #cloudfront
- Game Engine
| 일 | 월 | 화 | 수 | 목 | 금 | 토 |
|---|---|---|---|---|---|---|
| 1 | 2 | 3 | 4 | 5 | 6 | 7 |
| 8 | 9 | 10 | 11 | 12 | 13 | 14 |
| 15 | 16 | 17 | 18 | 19 | 20 | 21 |
| 22 | 23 | 24 | 25 | 26 | 27 | 28 |
글 보관함