[mapper_parsing_exception] when creating the same document with different type

[tim_wong]

I am importing a document twice into the same index but different types. However, I got a [mapper_parsing_exception] when I import the document the second time.

Below are my steps.

1. Create the template for the index.

   {
     "order": 0,
     "template": "stats-*",
     "settings": {
       "index": {
         "number_of_shards": "1",
         "number_of_replicas": "1",
         "refresh_interval": "5s"
       }
     },
     "mappings": {
       "_default_": {
         "include_in_all": false,
         "dynamic_templates": [
           {
             "metrics_fields": {
               "path_match": "metrics.*",
               "mapping": {
                 "scaling_factor": 100,
                 "type": "scaled_float"
               }
             }
           }
         ],
         "_all": {
           "norms": false,
           "enabled": true
         },
         "properties": {
           "@timestamp": {
             "type": "date"
           },
           "@version": {
             "type": "keyword"
           }
         }
       }
     },
     "aliases": {}
   }
   

2. Import the document into type a. The import was successful.

   POST stats-demo/a
   {
     "@timestamp": "2017-04-19T03:40:04.000Z",
     "@version": "1",
     "metrics": {
       "blockout": 0,
       "memory": 569.2,
       "cpu": 1420.4,
       "blockin": 123.5,
       "netout": 0,
       "memlimit": 1024,
       "netin": 0
     }
   }
   

3. Import the same document into type b of the same index. The following exception was thrown:

   {
     "error": {
       "root_cause": [
         {
           "type": "mapper_parsing_exception",
           "reason": "failed to parse"
         }
       ],
       "type": "mapper_parsing_exception",
       "reason": "failed to parse",
       "caused_by": {
         "type": "illegal_argument_exception",
         "reason": "Field [blockout] misses required parameter [scaling_factor]"
       }
     },
     "status": 400
   }
   

Below is the mapping of the stats-demo after the first successful import:

{
  "stats-demo": {
    "aliases": {},
    "mappings": {
      "_default_": {
        "include_in_all": false,
        "_all": {
          "enabled": true,
          "norms": false
        },
        "dynamic_templates": [
          {
            "metrics_fields": {
              "path_match": "metrics.*",
              "mapping": {
                "scaling_factor": 100,
                "type": "scaled_float"
              }
            }
          }
        ],
        "properties": {
          "@timestamp": {
            "type": "date"
          },
          "@version": {
            "type": "keyword"
          }
        }
      },
      "a": {
        "include_in_all": false,
        "_all": {
          "enabled": true,
          "norms": false
        },
        "dynamic_templates": [
          {
            "metrics_fields": {
              "path_match": "metrics.*",
              "mapping": {
                "scaling_factor": 100,
                "type": "scaled_float"
              }
            }
          }
        ],
        "properties": {
          "@timestamp": {
            "type": "date"
          },
          "@version": {
            "type": "keyword"
          },
          "metrics": {
            "properties": {
              "blockin": {
                "type": "scaled_float",
                "scaling_factor": 100
              },
              "blockout": {
                "type": "scaled_float",
                "scaling_factor": 100
              },
              "cpu": {
                "type": "scaled_float",
                "scaling_factor": 100
              },
              "memlimit": {
                "type": "scaled_float",
                "scaling_factor": 100
              },
              "memory": {
                "type": "scaled_float",
                "scaling_factor": 100
              },
              "netin": {
                "type": "scaled_float",
                "scaling_factor": 100
              },
              "netout": {
                "type": "scaled_float",
                "scaling_factor": 100
              }
            }
          }
        }
      }
    },
    "settings": {
      "index": {
        "refresh_interval": "5s",
        "number_of_shards": "1",
        "provided_name": "stats-demo",
        "creation_date": "1492669920734",
        "number_of_replicas": "1",
        "uuid": "_kbN2j5sTBieLc08Zcg6vA",
        "version": {
          "created": "5020299"
        }
      }
    }
  }
}

I am confusing why the same document couldn't be imported into another type of the same index.

Thank you for any help. Any possible solution is appreciated!

[warkolm (Mark Walkom)]

You cannot do this as it's a collision. You will need to put it in a new index or think of a different structure.

[tim_wong]

May I know how the collision happens? I thought that type is designed to store data with very similar data structure. I can actually import the same document twice if I am using the same type.

Thanks a lot!

[warkolm (Mark Walkom)]

Sorry, I don't know what I was thinking when I replied previously, but what I said is wrong!