Multi Search APIedit

Executes several searches with a single API request.

GET twitter/_msearch
{ }
{"query" : {"match" : { "message": "this is a test"}}}
{"index": "twitter2"}
{"query" : {"match_all" : {}}}

Requestedit

GET /<index>/_msearch

Descriptionedit

The multi search API executes several searches from a single API request. The format of the request is similar to the bulk API format and makes use of the newline delimited JSON (NDJSON) format.

The structure is as follows:

header\n
body\n
header\n
body\n

This structure is specifically optimized to reduce parsing if a specific search ends up redirected to another node.

The final line of data must end with a newline character \n. Each newline character may be preceded by a carriage return \r. When sending requests to this endpoint the Content-Type header should be set to application/x-ndjson.

Path parametersedit

<index>

(Optional, string) Comma-separated list or wildcard expression of index names used to limit the request.

To search all indices, use _all or omit this parameter.

Query parametersedit

allow_no_indices

(Optional, boolean) If true, the request does not return an error if a wildcard expression or _all value retrieves only missing or closed indices.

This parameter also applies to index aliases that point to a missing or closed index.

ccs_minimize_roundtrips
(Optional, boolean) If true, network roundtrips between the coordinating node and remote clusters are minimized for cross-cluster search requests. Defaults to true. See How cross-cluster search handles network delays.
expand_wildcards

(Optional, string) Controls what kind of indices that wildcard expressions can expand to. Valid values are:

all
Expand to open and closed indices.
open
Expand only to open indices.
closed
Expand only to closed indices.
none
Wildcard expressions are not accepted.

Defaults to open.

ignore_throttled
(Optional, boolean) If true, concrete, expanded or aliased indices are ignored when throttled. Defaults to false.
ignore_unavailable
(Optional, boolean) If true, missing or closed indices are not included in the response. Defaults to false.
max_concurrent_searches
(Optional, integer) Maximum number of concurrent searches the multi search API can execute.
max_concurrent_shard_requests

(Optional, integer) Maximum number of concurrent shard requests that each sub-search request executes per node. Defaults to 5.

You can use this parameter to prevent a request from overloading a cluster. For example, a default request hits all indices in a cluster. This could cause shard request rejections if the number of shards per node is high.

In certain scenarios, parallelism isn’t achieved through concurrent requests. In those cases, a low value in this parameter could result in poor performance. For example, in an environment where a very low number of concurrent search requests are expected, a higher value in this parameter may improve performance.

pre_filter_shard_size
(Optional, integer) Defines a threshold that enforces a pre-filter roundtrip to prefilter search shards based on query rewriting if the number of shards the search request expands to exceeds the threshold. This filter roundtrip can limit the number of shards significantly if for instance a shard can not match any documents based on it’s rewrite method i.e., if date filters are mandatory to match but the shard bounds and the query are disjoint. Defaults to 128.
rest_total_hits_as_int
(Optional, boolean) If true, hits.total are returned as an integer in the response. Defaults to false, which returns an object.
routing
(Optional, string) Custom routing value used to route search operations to a specific shard.
search_type

(Optional, string) Indicates whether global term and document frequencies should be used when scoring returned documents.

Options are:

query_then_fetch
(default) Documents are scored using local term and document frequencies for the shard. This is usually faster but less accurate.
dfs_query_then_fetch
Documents are scored using global term and document frequencies across all shards. This is usually slower but more accurate.
typed_keys
(Optional, boolean) Specifies whether aggregation and suggester names should be prefixed by their respective types in the response.

Request bodyedit

The request body contains a newline-delimited list of search <header> and search <body> objects.

<header>

(Required, object) Contains parameters used to limit or change the subsequent search body request.

This object is required for each search body but can be empty ({}) or a blank line.

allow_no_indices

(Optional, boolean) If true, the request does not return an error if a wildcard expression or _all value retrieves only missing or closed indices.

This parameter also applies to index aliases that point to a missing or closed index.

expand_wildcards

(Optional, string) Controls what kind of indices that wildcard expressions can expand to. Valid values are:

all
Expand to open and closed indices.
open
Expand only to open indices.
closed
Expand only to closed indices.
none
Wildcard expressions are not accepted.

Defaults to open.

ignore_unavailable
(Optional, boolean) If true, documents from missing or closed indices are not included in the response. Defaults to false.
index
(Optional, string or array of strings) Index name or alias used to limit the request. Wildcard expressions are supported. You can specify multiple indices as an array.
preference
(Optional, string) Node or shard used to perform the search. Random by default.
request_cache
(Optional, boolean) If true, the request cache can be used for this search. Defaults to index-level settings. See Shard request cache.
routing
(Optional, string) Custom routing value used to route search operations to a specific shard.
search_type

(Optional, string) Indicates whether global term and document frequencies should be used when scoring returned documents.

Options are:

query_then_fetch
(default) Documents are scored using local term and document frequencies for the shard. This is usually faster but less accurate.
dfs_query_then_fetch
Documents are scored using global term and document frequencies across all shards. This is usually slower but more accurate.
<body>

(Optional, object) Contains parameters for a search request:

aggregations
(Optional, aggregation object) Aggregations you wish to run during the search. See Aggregations.
query
(Optional, query DSL object) Query you wish to run during the search. Hits matching this query are returned in the response.
from
(Optional, integer) Starting offset for returned hits. Defaults to 0.
size
(Optional, integer) Number of hits to return. Defaults to 10.

Response bodyedit

responses
(array) Includes the search response and status code for each search request matching its order in the original multi search request. If there was a complete failure for a specific search request, an object with error message and corresponding status code will be returned in place of the actual search response.

Examplesedit

The header part includes which index / indices to search on, the search_type, preference, and routing. The body includes the typical search body request (including the query, aggregations, from, size, and so on).

$ cat requests
{"index" : "test"}
{"query" : {"match_all" : {}}, "from" : 0, "size" : 10}
{"index" : "test", "search_type" : "dfs_query_then_fetch"}
{"query" : {"match_all" : {}}}
{}
{"query" : {"match_all" : {}}}

{"query" : {"match_all" : {}}}
{"search_type" : "dfs_query_then_fetch"}
{"query" : {"match_all" : {}}}
$ curl -H "Content-Type: application/x-ndjson" -XGET localhost:9200/_msearch --data-binary "@requests"; echo

Note, the above includes an example of an empty header (can also be just without any content) which is supported as well.

The endpoint allows to also search against an index/indices in the URI itself, in which case it will be used as the default unless explicitly defined otherwise in the header. For example:

GET twitter/_msearch
{}
{"query" : {"match_all" : {}}, "from" : 0, "size" : 10}
{}
{"query" : {"match_all" : {}}}
{"index" : "twitter2"}
{"query" : {"match_all" : {}}}

The above will execute the search against the twitter index for all the requests that don’t define an index, and the last one will be executed against the twitter2 index.

The search_type can be set in a similar manner to globally apply to all search requests.

Template supportedit

Much like described in Search Template for the _search resource, _msearch also provides support for templates. Submit them like follows for inline templates:

GET _msearch/template
{"index" : "twitter"}
{ "source" : "{ \"query\": { \"match\": { \"message\" : \"{{keywords}}\" } } } }", "params": { "query_type": "match", "keywords": "some message" } }
{"index" : "twitter"}
{ "source" : "{ \"query\": { \"match_{{template}}\": {} } }", "params": { "template": "all" } }

You can also create search templates:

POST /_scripts/my_template_1
{
    "script": {
        "lang": "mustache",
        "source": {
            "query": {
                "match": {
                    "message": "{{query_string}}"
                }
            }
        }
    }
}
POST /_scripts/my_template_2
{
    "script": {
        "lang": "mustache",
        "source": {
            "query": {
                "term": {
                    "{{field}}": "{{value}}"
                }
            }
        }
    }
}

You can use search templates in a _msearch:

GET _msearch/template
{"index" : "main"}
{ "id": "my_template_1", "params": { "query_string": "some message" } }
{"index" : "main"}
{ "id": "my_template_2", "params": { "field": "user", "value": "test" } }

Partial responsesedit

To ensure fast responses, the multi search API will respond with partial results if one or more shards fail. See Shard failures for more information.