Aggregation over big data

[Filip_Sirc (Filip Širc)]

Hello,

i would like to ask you for advice.
I have index with milions of documents and need to query it (query will return cca. 100 000 documents) and use results for aggregations.

My current code is:

{
	"query": {
		"terms": {
			"status": ["NEW", "READY"]
		}
	},
	"aggregations": {
		"type": {
			"terms": {
				"field": "type",
				"size": 10000,
			},
			"aggregations": {
				"subtype": {
					"terms": {
						"field": "subtype",
						"size": 100
					},
					"aggregations": {
						"firstDate": {
							"min": {
								"field": "date"
							}
						},
						"lastDate": {
							"max": {
								"field": "date"
							}
						}
					}
				}
			}
		}
	}
}

Problem is, that the aggregations and documentSizes in final buckets are incomplete and thus aggregations incorrect.
I guess the reason is the limit 10 000 documents per query.
I am not owner of the database so I can't change the limit.

Is there any posibility how to get correct results?

Thank you!
Filip