I'm getting json_parse_exception on reindex from a bunch of daily indices to a consolidated monthly one, quite likely due to the same field in different daily indices being automapped to different types.
How do I tell reindex to ignore these errors - drop the erroneous fields, drop the records that contain them, I don't care really, it's just that I do want the other 99.99% of the data?
{
"took": 6943,
"timed_out": false,
"total": 16776617,
"updated": 6999,
"created": 0,
"deleted": 0,
"batches": 7,
"version_conflicts": 0,
"noops": 0,
"retries": {
"bulk": 0,
"search": 0
},
"throttled_millis": 0,
"requests_per_second": -1,
"throttled_until_millis": 0,
"failures": [
{
"index": "filebeat-2018.01",
"type": "doc",
"id": "AWEDf0pgq8giAw7TVsri",
"cause": {
"type": "mapper_parsing_exception",
"reason": "failed to parse [json.provision.assets.asset-instance.virtual]",
"caused_by": {
"type": "json_parse_exception",
"reason": "Current token (VALUE_TRUE) not numeric, can not use numeric value accessors\n at [Source: org.elasticsearch.common.bytes.BytesReference$MarkSupportingStreamInputWrapper@5b323cb5; line: 1, column: 596]"
}
},
"status": 400
}
]
}