> ## Documentation Index
> Fetch the complete documentation index at: https://docs.appsignal.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Host metrics

The AppSignal agent collects various system metrics, which allows you to correlate performance issues and errors to abnormal host metrics. This data is available in the [Host metrics](https://appsignal.com/redirect-to/app?to=host_metrics) section in the app overview, which allows you to inspect and [compare multiple hosts](https://blog.appsignal.com/2018/02/20/comparing-hosts.html). We also show a host metrics overview on the sample detail page for error and performance incidents.

For a preview of how host metrics look in the AppSignal interface, please see our [host metrics](https://www.appsignal.com/tour/hosts) tour page.

<Tip>
  **Note**: This feature is available in these packages:

  * Ruby: gem version `1.2` and newer. Enabled by default since version `1.3`.
  * Elixir: package version `0.10.0` and newer. Enabled by default since
    version `0.10.0`.
  * Node.js: all package versions.
  * JavaScript for Front-end: none package versions.
</Tip>

<Tip>
  **Note**: This feature is not available on the following architectures:

  * macOS/OSX (`darwin`)
  * FreeBSD

  A list of supported Operating Systems is available on the [Supported
  Operating Systems](/support/operating-systems) page.
</Tip>

## Collected host metrics

The following host metrics are collected by the AppSignal agent for every minute on your system.

| Metric          | Description                                                                                                                                                                               |
| --------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| CPU usage       | User, nice, system, idle and iowait in percentages. <br />Read more about [CPU metrics](https://blog.appsignal.com/2018/03/06/understanding-cpu-statistics.html)  in our academy article. |
| Load average    | 1 minute load average on the host.                                                                                                                                                        |
| Memory usage    | Available, free and used memory. Also includes swap total and swap used.                                                                                                                  |
| Disk usage      | Percentage of every disk used.                                                                                                                                                            |
| Disk IO         | Throughput of data read from and written to every disk.                                                                                                                                   |
| Network traffic | Throughput of data received and transmitted through every network interface.                                                                                                              |

These host metrics are collected by default. To disable it, use the `enable_host_metrics` configuration option, for:

* [Ruby](/ruby/configuration/options#option-enable_host_metrics)
* [Elixir](/elixir/configuration/options#option-enable_host_metrics)
* [Node.js](/nodejs/3.x/configuration/options#option-enable_host_metrics)
* [Python](/python/configuration/options#option-enable_host_metrics)
* [Collector](/collector/configuration/options#option-enable_host_metrics) for [OpenTelemetry](/opentelemetry)
* [Standalone agent](/standalone-agent/configuration/options#option-enable_host_metrics)

### Environment metadata

The AppSignal agent reports the following information of the host. Once received this metadata is shown on the [host metrics index page](https://appsignal.com/redirect-to/app?to=host_metrics) per host to provide more detail of all hosts running an app. Use this metadata to spot any differences between hosts that could cause differences in behavior.

* Host architecture, 32 or 64-bit.
* Operating System, either Linux, macOS or Windows.
* Operating System Distribution, for Linux this reports Ubuntu, Fedora, etc.
* Operating System version, the version of the Operating System.
* Kernel version, for Linux this reports the installed kernel version.

This metadata is collected by default. To disable it, use the `send_environment_metadata` configuration option, for:

* [Ruby](/ruby/configuration/options#option-send_environment_metadata)
* [Elixir](/elixir/configuration/options#option-send_environment_metadata)
* [Node.js](/nodejs/3.x/configuration/options#option-send_environment_metadata)
* [Python](/python/configuration/options#option-send_environment_metadata)

## Heroku support

To use host metrics on Heroku, head to the [Heroku host metrics][heroku support] page.

## Docker/container support {/* id: container-support */}

To use host metrics on (Docker) containers, head to the [container host metrics](/metrics/host-metrics/containers)
page.

## Dokku support

<Tip>
  **Note**: Dokku support is available in packages Ruby gem 3.0.13, Elixir
  package 2.2.0, Node.js package 2.0.0 and newer.
</Tip>

[Dokku](https://github.com/dokku/dokku) has host metrics enabled by default as long as your Dokku application has the `DOKKU_ROOT` environment variable set up.

[heroku support]: /heroku/host-metrics.html

[container support]: /metrics/host-metrics/containers.html