[[java-rest-high-x-pack-ml-start-datafeed]]
=== Start Datafeed API

The Start Datafeed API provides the ability to start a {ml} datafeed in the cluster.
It accepts a `StartDatafeedRequest` object and responds
with a `StartDatafeedResponse` object.

[[java-rest-high-x-pack-ml-start-datafeed-request]]
==== Start Datafeed Request

A `StartDatafeedRequest` object is created referencing a non-null `datafeedId`.
All other fields are optional for the request.

["source","java",subs="attributes,callouts,macros"]
--------------------------------------------------
include-tagged::{doc-tests}/MlClientDocumentationIT.java[x-pack-ml-start-datafeed-request]
--------------------------------------------------
<1> Constructing a new request referencing an existing `datafeedId`

==== Optional Arguments

The following arguments are optional.

["source","java",subs="attributes,callouts,macros"]
--------------------------------------------------
include-tagged::{doc-tests}/MlClientDocumentationIT.java[x-pack-ml-start-datafeed-request-options]
--------------------------------------------------
<1> Set when the datafeed should end, the value is exclusive.
May be an epoch seconds, epoch millis or an ISO 8601 string.
"now" is a special value that indicates the current time.
If you do not specify an end time, the datafeed runs continuously.
<2> Set when the datafeed should start, the value is inclusive.
May be an epoch seconds, epoch millis or an ISO 8601 string.
If you do not specify a start time and the datafeed is associated with a new job,
the analysis starts from the earliest time for which data is available.
<3> Set the timeout for the request

[[java-rest-high-x-pack-ml-start-datafeed-execution]]
==== Execution

The request can be executed through the `MachineLearningClient` contained
in the `RestHighLevelClient` object, accessed via the `machineLearningClient()` method.

["source","java",subs="attributes,callouts,macros"]
--------------------------------------------------
include-tagged::{doc-tests}/MlClientDocumentationIT.java[x-pack-ml-start-datafeed-execute]
--------------------------------------------------
<1> Did the datafeed successfully start?

[[java-rest-high-x-pack-ml-start-datafeed-execution-async]]
==== Asynchronous Execution

The request can also be executed asynchronously:

["source","java",subs="attributes,callouts,macros"]
--------------------------------------------------
include-tagged::{doc-tests}/MlClientDocumentationIT.java[x-pack-ml-start-datafeed-execute-async]
--------------------------------------------------
<1> The `StartDatafeedRequest` to execute and the `ActionListener` to use when
the execution completes

The method does not block and returns immediately. The passed `ActionListener` is used
to notify the caller of completion. A typical `ActionListener` for `StartDatafeedResponse` may
look like

["source","java",subs="attributes,callouts,macros"]
--------------------------------------------------
include-tagged::{doc-tests}/MlClientDocumentationIT.java[x-pack-ml-start-datafeed-listener]
--------------------------------------------------
<1> `onResponse` is called back when the action is completed successfully
<2> `onFailure` is called back when some unexpected error occurs