---
title: Disk
description: The disk check gathers metrics on mounted disks.
breadcrumbs: Docs > Integrations > Disk
---

# Disk
Supported OS Integration version7.4.0
## Overview{% #overview %}

Collect metrics related to disk usage and IO.

**Minimum Agent version:** 6.0.0

## Setup{% #setup %}

### Installation{% #installation %}

The disk check is included in the [Datadog Agent](https://app.datadoghq.com/account/settings/agent/latest) package, so you don't need to install anything else on your server.

### Configuration{% #configuration %}

The Disk check is enabled by default, and the Agent collects metrics on all local partitions. To configure the check with custom options, edit the `disk.d/conf.yaml` file, in the `conf.d/` folder at the root of your [Agent's configuration directory](https://docs.datadoghq.com/agent/guide/agent-configuration-files/#agent-configuration-directory). See the [sample disk.d/conf.yaml](https://github.com/DataDog/integrations-core/blob/master/disk/datadog_checks/disk/data/conf.yaml.default) for all available configuration options.

#### Note for Windows hosts{% #note-for-windows-hosts %}

See the following scenarios where the Disk check can be used:

1. Monitoring physical drives

Monitoring physical drives that are represented by a disk letter (for example. C:, D:, etc.) is supported out of the box by the disk check without any special considerations.
Monitoring nested mount points
Monitoring mounted folders within a filesystem requires Administrator permissions. This is because the underlying Windows function call [FindFirstVolumeMountPoint](https://docs.microsoft.com/en-us/windows/win32/api/winbase/nf-winbase-findfirstvolumemountpointw) requires administrative permissions. To collect those metrics without granting Administrator permissions to the Agent, use the [PDH check](https://docs.datadoghq.com/integrations/pdh_check/#pagetitle) to collect mount point metrics from the corresponding perf counters.
Monitoring file shares
Collecting mount point metrics for file shares on Windows is only supported by using the `create_mounts` option in the configuration. On Windows, each mounted folder is only visible to the user who mounted the share. Therefore, the `create_mounts` option allows the Agent to create the mount points to monitor in the context of the Agent's user.
Monitoring Cluster Shared Volumes
Cluster Shared Volumes (CSVs) in Windows Failover Clustering require elevated access for proper discovery. While there may be other ways to grant the necessary permissions, running the Datadog Agent as LocalSystem is the most reliable and consistent method to ensure full CSV visibility and metric collection.

### Validation{% #validation %}

[Run the Agent's `status` subcommand](https://docs.datadoghq.com/agent/guide/agent-commands/#agent-status-and-information) and look for `disk` under the Checks section.

## Data Collected{% #data-collected %}

### Metrics{% #metrics %}

|  |
|  |
| **system.disk.free**(gauge)            | The amount of disk space that is free.*Shown as byte*                           |
| **system.disk.in\_use**(gauge)         | The amount of disk space in use as a percent of the total.*Shown as fraction*   |
| **system.disk.read\_time**(count)      | The time in ms spent reading per device*Shown as millisecond*                   |
| **system.disk.read\_time\_pct**(rate)  | Percent of time spent reading from disk.*Shown as percent*                      |
| **system.disk.total**(gauge)           | The total amount of disk space.*Shown as byte*                                  |
| **system.disk.used**(gauge)            | The amount of disk space in use.*Shown as byte*                                 |
| **system.disk.utilized**(gauge)        | The amount of disk space in use as a percentage of the total.*Shown as percent* |
| **system.disk.write\_time**(count)     | The time in ms spent writing per device.*Shown as millisecond*                  |
| **system.disk.write\_time\_pct**(rate) | Percent of time spent writing to disk.*Shown as percent*                        |
| **system.fs.inodes.free**(gauge)       | The number of free inodes.*Shown as inode*                                      |
| **system.fs.inodes.in\_use**(gauge)    | The number of inodes in use as a percent of the total.*Shown as fraction*       |
| **system.fs.inodes.total**(gauge)      | The total number of inodes.*Shown as inode*                                     |
| **system.fs.inodes.used**(gauge)       | The number of inodes in use.*Shown as inode*                                    |
| **system.fs.inodes.utilized**(gauge)   | The number of inodes in use as a percentage of the total.*Shown as percent*     |

### Events{% #events %}

The Disk check does not include any events.

### Service Checks{% #service-checks %}

**disk.read\_write**

Returns `CRITICAL` if filesystem is in read-only mode.

**Note**: This service check is not enabled by default. To enable the service check, set `service_check_rw` to `true` in the disk check's `conf.yaml` file.

*Statuses: ok, critical, unknown*

## Troubleshooting{% #troubleshooting %}

Need help? Contact [Datadog support](https://docs.datadoghq.com/help/).