Performance when using `time_zone` in date_historgram

[orweinberger]

I'm experiencing some weird performance results in Elasticsearch 5.2.0.

I've been trying to debug why its taking so long to load the discover page in Kibana, the default query and aggregation (for the top bar chart) takes about 4 seconds when looking at the query statistics in Kibana.

If I copy the query and run it in the console, I can see the same results. This is the query:

POST index-2017.11.16/_search
{
  "size": 500,
  "sort": [
    {
      "@timestamp": {
        "order": "desc",
        "unmapped_type": "boolean"
      }
    }
  ],
  "query": {
    "bool": {
      "must": [
        {
          "query_string": {
            "analyze_wildcard": true,
            "query": "*"
          }
        },
        {
          "range": {
            "@timestamp": {
              "gte": 1510783200000,
              "lte": 1510869599999,
              "format": "epoch_millis"
            }
          }
        }
      ],
      "must_not": []
    }
  },
  "_source": {
    "excludes": []
  },
  "aggs": {
    "2": {
      "date_histogram": {
        "field": "@timestamp",
        "interval": "30m",
        "time_zone": "Asia/Amman",
        "min_doc_count": 1
      }
    }
  },
  "stored_fields": [
    "*"
  ],
  "script_fields": {},
  "docvalue_fields": []
}

However, if I remove the time_zone attribute from the date_histogram, I get results back in 900ms. Is this a known issue? What causes this?

[orweinberger]

Any ideas?

[jpountz (Adrien Grand)]

Unfortunately, this is expected: time zones make it hard to identify the right bucket, especially for time zones that implement daylight saving time.