On 5/11/2015 6:13 AM, Bernd Fehling wrote: > Caused by: java.lang.IllegalArgumentException: Document contains at least one > immense term > in field="f_dcperson" (whose UTF8 encoding is longer than the max length > 32766), all of which were skipped. > Please correct the analyzer to not produce such terms. The prefix of the > first immense > term is: '[102, 111, 114, 32, 97, 32, 114, 101, 118, 105, 101, 119, 32, 115, > 101, 101, 32, 66, 114, > 111, 119, 110, 105, 110, 103, 32, 32, 32, 50, 48]...', original message: > bytes can be at most 32766 in length; got 38177 > at > org.apache.lucene.index.DefaultIndexingChain$PerField.invert(DefaultIndexingChain.java:687) > ...
The field in question is f_dcperson, which according to your schema is a "string" type. If your schema follows the example fieldType definitions, then "string" is a solr.StrField, where the entire input is treated as one term. The field is multiValued and a copyField destination, so each value that is sent is one term. I went looking for this message in the code. It is logged when a MaxBytesLengthExceededException is thrown. This error is complaining that the size of the *term* (since it's a "string" type, likely the contents of an individual copyField source field) you are sending to the f_dcperson field has exceeded 32766, which is apparently the largest size for that field type. You'll either need to fix your source data or pick a field type that can handle more data. Thanks, Shawn
