Do terms queries behave differently with runtime mappings?

Elastic Stack Elasticsearch
1

I have a query:

{
    "query": { "bool": { "filter": [
        { "terms": { "foo": ... (1000 terms) } },
        { "terms": { "bar": ... (1000 terms) } }
    ] } }
}

which works just fine as is. But if I run those terms off a runtime mapping...

{
    "runtime_mappins": {
        "foo-a": { "type": "keyword", "script": "emit(doc['foo'].value)" },
        "bar-a": { "type": "keyword", "script": "emit(doc['bar'].value)" },
    },
    "query": { "bool": { "filter": [
        { "terms": { "foo-a": ... (1000 terms) } },
        { "terms": { "bar-a": ... (1000 terms) } }
    ] } }
}

it fails with lots of

            {
                "type": "too_many_nested_clauses",
                "reason": "too_many_nested_clauses: Query contains too many nested clauses; maxClauseCount is set to 1024"
            },

Anyone have a clue why? It should only be 2 clauses - and without the runtime mapping, it seems to behave as such; it's only with the runtime mapping it complains

2

This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.