https://github.com/mevdschee/php-observability

High frequency logging in PHP and aggregating into metrics using Go
https://github.com/mevdschee/php-observability

logging metrics observability php prometheus

Last synced: 3 months ago
JSON representation

High frequency logging in PHP and aggregating into metrics using Go

• Host: GitHub
• URL: https://github.com/mevdschee/php-observability
• Owner: mevdschee
• License: mit
• Created: 2024-09-17T09:14:25.000Z (5 months ago)
• Default Branch: main
• Last Pushed: 2024-10-30T01:22:51.000Z (4 months ago)
• Last Synced: 2024-10-30T03:53:51.355Z (4 months ago)
• Topics: logging, metrics, observability, php, prometheus
• Language: Go
• Homepage:
• Size: 51.8 KB
• Stars: 6
• Watchers: 1
• Forks: 1
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# php-observability

A code base to showcase high frequency logging in PHP and aggregating into metrics using Go. Read the blog post:

[https://tqdev.com/2024-high-frequency-metrics-in-php-using-tcp-sockets](https://tqdev.com/2024-high-frequency-metrics-in-php-using-tcp-sockets)

and

[https://tqdev.com/2024-distributed-metrics-in-php-using-go-and-gob](https://tqdev.com/2024-distributed-metrics-in-php-using-go-and-gob)

### Requirements

- PHP 8 or higher (see special PHP 5/7 compatible version)

- Go 1.22 build environment

### Usage

To run the server:

    go run .

In bash run:

    for run in {1..100}; do php writer.php & done

And to stop:

    killall php

Now observe the metrics:

http://localhost:8080/

NB: The metrics are Prometheus compatible and follow the [OpenMetrics specification](https://github.com/OpenObservability/OpenMetrics/).

The metrics are also served in the Gob binary format on:

http://localhost:9999/

This endpoint can be scraped by other tools such as [metrics-db-importer](https://github.com/mevdschee/metrics-db-importer) or this tool itself to aggregate the metrics of multiple nodes (using the "-scrape" parameter).

### Example metrics

Here is an example of published metrics:

    # HELP database_calls_seconds A summary of the database calls.

    # TYPE database_calls_seconds summary

    database_calls_seconds_count{file="src/Controller/UserController.php@L123"} 6630

    database_calls_seconds_sum{file="src/Controller/UserController.php@L123"} 947.142

    # HELP database_calls_total_seconds A histogram of the database calls.

    # TYPE database_calls_total_seconds histogram

    database_calls_total_seconds_bucket{le="0.005"} 0

    database_calls_total_seconds_bucket{le="0.01"} 0

    database_calls_total_seconds_bucket{le="0.025"} 0

    database_calls_total_seconds_bucket{le="0.05"} 0

    database_calls_total_seconds_bucket{le="0.1"} 0

    database_calls_total_seconds_bucket{le="0.25"} 6630

    database_calls_total_seconds_bucket{le="0.5"} 6630

    database_calls_total_seconds_bucket{le="1"} 6630

    database_calls_total_seconds_bucket{le="2.5"} 6630

    database_calls_total_seconds_bucket{le="5"} 6630

    database_calls_total_seconds_bucket{le="10"} 6630

    database_calls_total_seconds_bucket{le="+Inf"} 6630

    database_calls_total_seconds_sum 947.142

    database_calls_total_seconds_count 6630

Enjoy!