---
title: PHP FPM
description: Monitor process states, slow requests, and accepted requests.
breadcrumbs: Docs > Integrations > PHP FPM
---

# PHP FPM
Supported OS Integration version6.3.0


## Overview{% #overview %}

The PHP-FPM check monitors the state of your FPM pool and tracks request performance.

**Minimum Agent version:** 6.0.0

## Setup{% #setup %}

### Installation{% #installation %}

The PHP-FPM check is included in the [Datadog Agent](https://app.datadoghq.com/account/settings/agent/latest) package. No additional installation is needed on your server.

### Configuration{% #configuration %}

Follow the instructions below to configure this check for an Agent running on a host. For containerized environments, see the Containerized section.

{% tab title="Host" %}
#### Host{% #host %}

To configure this check for an Agent running on a host:

1. Edit the `php_fpm.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 php_fpm.d/conf.yaml](https://github.com/DataDog/integrations-core/blob/master/php_fpm/datadog_checks/php_fpm/data/conf.yaml.example) for all available configuration options:

   ```yaml
   init_config:
   
   instances:
     ## @param status_url - string - required
     ## Get metrics from your FPM pool with this URL
     ## The status URLs should follow the options from your FPM pool
     ## See http://php.net/manual/en/install.fpm.configuration.php
     ##   * pm.status_path
     ## You should configure your fastcgi passthru (nginx/apache) to catch these URLs and
     ## redirect them through the FPM pool target you want to monitor (FPM `listen`
     ## directive in the config, usually a UNIX socket or TCP socket.
     #
     - status_url: http://localhost/status
   
       ## @param ping_url - string - required
       ## Get a reliable service check of your FPM pool with `ping_url` parameter
       ## The ping URLs should follow the options from your FPM pool
       ## See http://php.net/manual/en/install.fpm.configuration.php
       ##   * ping.path
       ## You should configure your fastcgi passthru (nginx/apache) to
       ## catch these URLs and redirect them through the FPM pool target
       ## you want to monitor (FPM `listen` directive in the config, usually
       ## a UNIX socket or TCP socket.
       #
       ping_url: http://localhost/ping
   
       ## @param use_fastcgi - boolean - required - default: false
       ## Communicate directly with PHP-FPM using FastCGI
       #
       use_fastcgi: false
   
       ## @param ping_reply - string - required
       ## Set the expected reply to the ping.
       #
       ping_reply: pong
   ```

1. [Restart the Agent](https://docs.datadoghq.com/agent/guide/agent-commands/#start-stop-and-restart-the-agent).

{% /tab %}

{% tab title="Containerized" %}
#### Containerized{% #containerized %}

For containerized environments, see the [Autodiscovery Integration Templates](https://docs.datadoghq.com/agent/kubernetes/integrations/) for guidance on applying the parameters below.

| Parameter            | Value                                                                                                                    |
| -------------------- | ------------------------------------------------------------------------------------------------------------------------ |
| `<INTEGRATION_NAME>` | `php_fpm`                                                                                                                |
| `<INIT_CONFIG>`      | blank or `{}`                                                                                                            |
| `<INSTANCE_CONFIG>`  | `{"status_url":"http://%%host%%/status", "ping_url":"http://%%host%%/ping", "use_fastcgi": false, "ping_reply": "pong"}` |

#### Extras{% #extras %}

##### Multiple pools{% #multiple-pools %}

It is possible to monitor multiple PHP-FPM pools using the same proxy server, a common scenario when running on Kubernetes. To do so, modify your server's routes to point to different PHP-FPM instances. Here is an example NGINX configuration:

```text
server {
    ...

    location ~ ^/(status1|ping1)$ {
        access_log off;
        fastcgi_pass instance1_ip:instance1_port;
        include fastcgi_params;
        fastcgi_param SCRIPT_FILENAME $document_root$fastcgi_script_name;
    }

    location ~ ^/(status2|ping2)$ {
        access_log off;
        fastcgi_pass instance2_ip:instance2_port;
        include fastcgi_params;
        fastcgi_param SCRIPT_FILENAME $document_root$fastcgi_script_name;
    }
}
```

If you find this approach too tedious at scale, setting `use_fastcgi` to `true` instructs the check to bypass any proxy servers and communicate directly with PHP-FPM using FastCGI. The default port is `9000` for when omitted from `status_url` or `ping_url`.

##### Unix sockets{% #unix-sockets %}

If your PHP-FPM installation uses unix sockets, you have to use the below syntax for `status_url`, `ping_url` and enable `use_fastcgi`:

| Parameter     | Value                             |
| ------------- | --------------------------------- |
| `status_url`  | `unix:///<FILE_PATH>.sock/status` |
| `ping_url`    | `unix:///<FILE_PATH>.sock/ping`   |
| `ping_reply`  | `pong`                            |
| `use_fastcgi` | `true`                            |

**Note**: With Autodiscovery, if the Agent runs in a separate container/task/pod, it doesn't have access to the Unix sockets file of your FPM pool. It order to address this, run the Agent as a sidecar.
{% /tab %}

### Validation{% #validation %}

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

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

### Metrics{% #metrics %}

|  |
|  |
| **php\_fpm.listen\_queue.size**(gauge)     | Size of the socket queue of pending connections                                                                 |
| **php\_fpm.processes.active**(gauge)       | Total number of active processes*Shown as process*                                                              |
| **php\_fpm.processes.idle**(gauge)         | Total number of idle processes*Shown as process*                                                                |
| **php\_fpm.processes.max\_active**(count)  | The maximum number of processes that can run at once*Shown as process*                                          |
| **php\_fpm.processes.max\_reached**(count) | The number of times the process limit has been reached (when pm tries to start more children)*Shown as process* |
| **php\_fpm.processes.total**(gauge)        | Total number of processes*Shown as process*                                                                     |
| **php\_fpm.requests.accepted**(count)      | Total number of accepted requests*Shown as request*                                                             |
| **php\_fpm.requests.slow**(count)          | Total number of slow requests*Shown as request*                                                                 |

### Events{% #events %}

The PHP-FPM check does not include any events.

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

**php\_fpm.can\_ping**

Returns `CRITICAL` if the Agent is unable to get a pong from the PHP-FPM ping URL.

*Statuses: ok, critical*

## Troubleshooting{% #troubleshooting %}

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