StatsD Client Bundle

This bundle helps you to monitor your application.

It uses statsd-php-client as Client for the etsy/Statsd API.

It Includes

a Service, a Factory and an Entity callable by the DIC
a Monolog Handler
a series of Collectors in order to aggregate your personal data.
a series of Commands in order to execute data also from cli.

Simple usage!

$this->get('statsd')
     ->increment('log.error')
     ->gauge('log.rate', 25)
     ->flush();

Check out the documentation

Articles

Abstract

Monitoring what is happening in your production Symfony2 application, with StatsD and Graphite.

Within this bundle you could use ready-to-go monitor for

Monolog
Amount of visitor
Logged vs Anonymous user
SQL: Verbs used (amount of select/set/update/upgrade/delete) (this feature is not suggested, please use a server monitor tool eg. nagios, munin)
Amount of PHP ram used
You could create your own using the Factory and the Service.

On dev env the data are not pushed to the socket but to syslog.

If debug mode is enabled no packet are sent over the udp socket, and you could see what's going on reading the syslog with tail -f /var/log/syslog

Mar 19 00:48:11 liuggio nginx: statsd-open
Mar 19 00:48:11 liuggio nginx: statsd-write "tvision.visitor:1|tvision.memory:43520|tvision.user.anonymous:1|tvision.query.start:1|tvision.query.insert:1|c" 118 Bytes
Mar 19 00:48:11 liuggio nginx: statsd-close

Note: by default in dev environment debug is true.

One UDP packet to rule them all

With a proper config file, your StatsD server will receive in a single upd packet:

prefix.visitor:1|c                  # Increment visitor
prefix.user.anonymous:1|c           # The visitor is not logged

prefix.log.app.error.user_action:1  # Stats by monolog with $this->get('logger')->error('user_action')

prefix.memory:4480|g                # The php memory peak was 4480 MB

prefix.exception.exception.500:1|c  # Exception occurred, the Exception code was 500

prefix.query.insert:1|c             # Query Insert
prefix.query.select:1|c             # Query Select