Elasticsearch

Elasticsearch

1. Create index
curl -X PUT "http://<>:9200/index-name" -d '{
  "mappings": {
    "record": {
      "properties": {
        "actionStartedOn": {
          "type": "date"
        },
        "rddCount": {
          "type": "long"
        },
        "rdd_created": {
          "type": "long"
        },
        "rdd_finished": {
          "type": "long"
        },
        "rdd_started": {
          "type": "long"
        },
        "topology_created": {
          "type": "long"
        },
        "topology_finished": {
          "type": "long"
        },
        "topology_started": {
          "type": "long"
        },
        "topology_failed": {
          "type": "long"
        },
        "vis_command_finished": {
          "type": "long"
        },
        "vis_data_succeeded": {
          "type": "long"
        },
        "topology_cancelled": {
          "type": "long"
        },
        "rdd_cancelled": {
          "type": "long"
        },
        "vis": {
          "type": "string",
          "index": "not_analyzed"
        }
      }
    }
  }
}'

2. Get statistic
curl -X GET "http://<>:9200/index-name/_stats?pretty"

3. Delete
curl -X DELETE "http://<>:9200/index-name"

4. Get unique values
curl -X GET "http://<>:9200/index-name/_search" -d '{
   "size": 0,
   "aggs": {
     "langs": {
       "terms": {
         "field": "<field-name>"
       }
     }
   }
 }'
 
5. Update all documents by query (WHERE <query field name>=<query field value):
curl -X POST http://<>:9200/index-name/_update_by_query -d '{
"script": {
    "inline": "ctx._source.<field name>=\"new value\""
  },
  "query": {
    "term": {
      "<query field name>": "<query field value>"
    }
  }
}'

for this action inline script should be enabled:
  script.engine.groovy.inline.update: true
  script.engine.groovy.indexed.update: true
  script.inline: true
in
  config/elasticsearch.yml
 
6. Update field <fieldName> that is missing ("NULL")
curl -X POST http://<host>:9200/<index_name>/_update_by_query -d '{
  "script": {
    "inline": "ctx._source.<fieldName>=\"newValue\""
  },
  "query": {
    "bool": {
      "must_not": {
        "exists": {
          "field": "<fieldName>"
        }
      }
    }
  }
}'