# Agent management

## Agents

The **Agents** tab shows high-level information for each agent, including:

| **Field**                                    | **Description**                                                                              |
| -------------------------------------------- | -------------------------------------------------------------------------------------------- |
| Agent name                                   | Usually based on the host name.                                                              |
| Agent type                                   | OTel collector type, such as agent, gateway, or cluster collector.                           |
| OS                                           | The OS of the host.                                                                          |
| Agent status                                 | Indicates whether the agent has an active connection.                                        |
| Pipelines                                    | The telemetry pipelines configured for the agent, such as logs, metrics, traces or profiles. |
| Agent version                                | The version of the agent.                                                                    |
| Cluster                                      | The cluster where the agent is deployed.                                                     |
| Active configuration                         | The configuration assigned to this agent.                                                    |
| Agent CPU usage                              | Last 15 minutes CPU usage average.                                                           |
| Agent memory usage                           | Last 15 minutes memory usage average.                                                        |
| Records refused by the receiver on the agent | Number of records dropped by the agent's receiver.                                           |
| Send failures by the exporter on the agent   | Number of failures reported by the agent's exporter when sending data.                       |

### Search

Use the search bar to filter agents by key-value fields (Lucene syntax) or by free-text search across all fields. For example, searching for `gateway` returns agents with an agent type of `gateway` as well as agents with gateway in the name.

Note

Free-text search may return broad results. Searching specific fields produces the most precise matches.

### Group by

Use **Group by** to organize agents into nested groups based on attributes such as agent type, operating system, pipelines, or other labels. Grouping helps you review large fleets by clustering related agents. Each group shows the number of agents it includes. Groups can be expanded or collapsed for navigation. Grouping does not affect search or filtering and helps you spot patterns, compare agent subsets, and understand distribution across environments at a glance.

### Investigate a single agent

Select an agent to open a detailed view with several tabs:

- **Overview** - Live status, CPU and memory usage, and receiver or exporter errors.
- **Details** - A summary of the resource's identity, metadata, configuration, and labels from available data sources.
- **Relationships** - Significant infrastructure relationships that this agent has. For example, the EC2 host on which it is deployed.
- **Metrics** - Metrics the agent publishes through OpAMP.
- **Configuration** - The OpenTelemetry configuration for this agent

**Overview**

The Overview section provides high-level graphs that display "at a glance" view of the health of your agent.

At the top, the **Live Status** view gives aggregate views of four golden signals for your agent. Below, the **Key Metrics** section shows a time-series view of the same data.

**Details**

The Details section aggregates data from available sources, including OpenTelemetry, Kubernetes, and cloud providers, to provide a comprehensive view of how the resource is defined and configured in your environment.

**Relationships**

The Relationships tab shows the agent's dependencies in the infrastructure. This can help determine whether an issue originates from the agent or the underlying infrastructure.

Available relationship categories include:

- Pods
- Containers
- Namespaces
- Clusters
- Storage

Select any item to open its dedicated view to drill down further into the resource details.

**Metrics**

The Metrics tab lets you explore the metrics the agent produces. This can help identify performance issues or unusual behavior.

You can search for metrics by entering text in the search bar. For example, searching for `consumer` returns only metrics with `consumer` in the metric name.

To view the underlying query behind a metric, select the query icon in the top right corner of the metric panel.

**Configuration**

The Configuration tab provides a navigable view of the OpenTelemetry configuration for the agent. You can review specific configuration fields to investigate potential issues.

A search bar allows filtering for values within the configuration. Use the icons next to the search bar to copy the full configuration or download it as a YAML file.

See [Configuration management](https://coralogix.com/docs/user-guides/fleet-management/configuration-management/index.md) for more information about managing and updating agent configurations.

## Next steps

Visualize and validate your OpenTelemetry Collector configurations with [Config Navigator](https://coralogix.com/docs/user-guides/fleet-management/config-navigator/index.md).