SpringBootData ElasticSearch cannot create index on non-indexed field

Question

I'm struggling creating a Document with field that are specified to be analyzed by ElasticSearch. I'm using Spring Data Elasticsearch version 4.1.6.

This is my Entity

@Document(indexName = "employee-info")
@Data
@JsonIgnoreProperties(ignoreUnknown = true)
public class EmployeeInfo {
    @Id
    @Field(type = FieldType.Keyword)
    private String empNo;
    @Field(type = FieldType.Keyword)
    private String pinCd;
    @Field(type = FieldType.Text)
    private String empNm;
    @Field(index = false)
    private String fullDeptCd;//00082;N0001;N0002;N0060;LW00002289;LW00002384;LW00002501",

    // ... other fields

}

Upon Bootstrapping, it logs:

2022-03-09 13:03:22,477 WARN  [main] SimpleElasticsearchRepository.java:96: Cannot create index: Elasticsearch exception [type=mapper_parsing_exception, reason=No type specified for field [fullDeptCd]]; nested exception is ElasticsearchStatusException[Elasticsearch exception [type=mapper_parsing_exception, reason=No type specified for field [fullDeptCd]]]

and the mapping:

{
  "mappings": {
    "_doc": {
      "properties": {
        "_class": {
          "type": "text",
          "fields": {
            "keyword": {
              "type": "keyword",
              "ignore_above": 256
            }
          }
        },
        "empNo": {
          "type": "text",
          "fields": {
            "keyword": {
              "type": "keyword",
              "ignore_above": 256
            }
          }
        },
        "fullDeptCd": {
          "type": "text",
          "fields": {
            "keyword": {
              "type": "keyword",
              "ignore_above": 256
            }
          }
        },
        ...

• Why empNo is text but its fields are keyword and it also doesn't have index?
• Why fullDeptCd mapping is exactly the same as empNo?

What I understand:

• FieldType.Keyword: this field will not be tokenized, but it is still indexed by default.
• FieldType.Text: this field will be tokenized by the analyzer.
• index = false: do not analyzed and index this field.

I feel like it's very difficult to work with ElasticSearch with SpringBoot, a lot of property in @Field annotation are not documented and I don't know how to use them properly.

If you have experience with ElasticSearch, please help.

Thank you in advance.

Answer 1

Looks like for some reason, the default FieldType.Auto not working, I have to explicitly specify type for all fields.