[[configuring-elastic-agent]]
== Collecting {es} monitoring data with {agent}

[subs="attributes"]
++++
<titleabbrev>Collecting monitoring data with {agent}</titleabbrev>
++++

In 8.5 and later, you can use {agent} to collect data about {es} and ship it to
the monitoring cluster, rather than <<configuring-metricbeat,using {metricbeat}>>
or routing it through exporters as described in <<collecting-monitoring-data>>.

[discrete]
=== Prerequisites

* (Optional) Create a monitoring cluster as described in <<monitoring-production>>.

* Create a user on the production cluster that has the
`remote_monitoring_collector` {ref}/built-in-roles.html[built-in role].

[discrete]
=== Add {es} monitoring data

To collect {es} monitoring data, add an {es} integration to an {agent} and
deploy it to the host where {es} is running.

. Go to the {kib} home page and click **Add integrations**.
. In the query bar, search for and select the **{es}** integration for
{agent}.
. Read the overview to make sure you understand integration requirements and
other considerations.
. Click **Add Elasticsearch**.
+
TIP: If you're installing an integration for the first time, you may be prompted
to install {agent}. Click **Add integration only (skip agent installation)**.

. Configure the integration name and optionally add a description. Make sure you
configure all required settings:
.. Under **Collect Elasticsearch logs**, modify the log paths to match your {es}
environment.
.. Under **Collect Elasticsearch metrics**, make sure the hosts setting points to
your {es} host URLs. By default, the integration collects {es} monitoring
metrics from `localhost:9200`. If that host and port number are not correct,
update the `hosts` setting. If you configured {es} to use encrypted
communications, you must access it via HTTPS. For example, use a `hosts` setting
like `https://localhost:9200`.
.. Expand **Advanced options**. If the Elastic {security-features} are enabled,
enter the username and password of a user that has the
`remote_monitoring_collector` role.
.. Specify the scope:
** Specify `cluster` if each entry in the hosts list indicates a single
endpoint for a distinct {es} cluster (for example, a load-balancing proxy
fronting the cluster that directs requests to the master-ineligible nodes in the
cluster).
** Otherwise, accept the default scope, `node`. If this scope is set, you
will need to install {agent} on each {es} node to collect all metrics. {agent}
will collect most of the metrics from the elected master of the cluster, so you
must scale up all your master-eligible nodes to account for this extra load. Do
not use this `node` if you have dedicated master nodes.
. Choose where to add the integration policy. Click **New hosts** to add it to
new agent policy or **Existing hosts** to add it to an existing agent policy.
. Click **Save and continue**. This step takes a minute or two to complete. When
it's done, you'll have an agent policy that contains an integration for
collecting monitoring data from {es}.
. If an {agent} is already assigned to the policy and deployed to the host where
{es} is running, you're done. Otherwise, you need to deploy an {agent}. To
deploy an {agent}:
.. Go to **{fleet} -> Agents**, then click **Add agent**.
.. Follow the steps in the **Add agent** flyout to download, install,
and enroll the {agent}. Make sure you choose the agent policy you created
earlier.
. Wait a minute or two until incoming data is confirmed.
. {kibana-ref}/monitoring-data.html[View the monitoring data in {kib}].