Anomaly detection fails with Document contains at least one immense term in field="model.normalized_keyword" (whose UTF8 encoding is longer than the max length 32766), all of which were skipped

Plugins Machine Learning
1

Hi All,
I am trying to create an anomaly detection model on a simple index with around 5 float fields.
On initialization of the detector, it fails in the UI. The detailed logger error is
Document contains at least one immense term in field=“model.normalized_keyword” (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: ‘[114, 110, 103, 32, 100, 105, 109, 101, 110, 115, 105, 111, 110, 115, 32, 49, 32, 115, 97, 109, 112, 108, 101, 115, 105, 122, 101, 32, 50, 53]…’, original message: bytes can be at most 32766 in length; got 1442859. How do I resolve this error? Thanks in advance

2

hi, @Harishram , sorry to reply so late, can you share your detector configuration and the data source index mapping? And which ODFE/OpenSearch version are you running?