Kohana-Cron

This module provides a way to schedule tasks (jobs) within your Kohana application.

Installation

Step 1: Download the module into your modules subdirectory.

Step 2: Enable the module in your bootstrap file:

/**
 * Enable modules. Modules are referenced by a relative or absolute path.
 */
Kohana::modules(array(
	'cron'       => MODPATH.'cron',
	// 'auth'       => MODPATH.'auth',       // Basic authentication
	// 'codebench'  => MODPATH.'codebench',  // Benchmarking tool
	// 'database'   => MODPATH.'database',   // Database access
	// 'image'      => MODPATH.'image',      // Image manipulation
	// 'orm'        => MODPATH.'orm',        // Object Relationship Mapping
	// 'pagination' => MODPATH.'pagination', // Paging of results
	// 'userguide'  => MODPATH.'userguide',  // User guide and API documentation
));

Step 3: Make sure the settings in config/cron.php are correct for your environment. If not, copy the file to application/config/cron.php and change the values accordingly.

Usage

In its simplest form, a task is a PHP callback and times at which it should run. To configure a task call Cron::set($name, array($frequency, $callback)) where $frequency is a string of date and time fields identical to those found in crontab. For example,

Cron::set('reindex_catalog', array('@daily', 'Catalog::regenerate_index'));
Cron::set('calendar_notifications', array('*/5 * * * *', 'Calendar::send_emails'));

Configured tasks are run with their appropriate frequency by calling Cron::run(). Call this method in your bootstrap file, and you're done!

Advanced Usage

A task can also be an instance of Cron that extends next() and/or execute() as needed. Such a task is configured by calling Cron::set($name, $instance).

If you have access to the system crontab, you can run Cron less (or more) than once every request. You will need to modify the lines where the request is handled in your bootstrap file to prevent extraneous output. The default is:

/**
 * Execute the main request. A source of the URI can be passed, eg: $_SERVER['PATH_INFO'].
 * If no source is specified, the URI will be automatically detected.
 */
echo Request::instance()
	->execute()
	->send_headers()
	->response;

Change it to:

if ( ! defined('SUPPRESS_REQUEST'))
{
	/**
	 * Execute the main request. A source of the URI can be passed, eg: $_SERVER['PATH_INFO'].
	 * If no source is specified, the URI will be automatically detected.
	 */
	echo Request::instance()
		->execute()
		->send_headers()
		->response;
}

Then set up a system cron job to run your application's Cron once a minute:

* * * * * /usr/bin/php -f /path/to/kohana/modules/cron/run.php

The included run.php should work for most cases, but you are free to call Cron::run() in any way you see fit.

Multi-streaming and using groups

Some jobs takes a lot of time, for example, and some jobs you need to be done quick and with a high frequency.

Cron::set('reindex_catalog', array('@hourly', 'Catalog::regenerate_index')); // takes ~2 hours
Cron::set('update_catalog_prices', array('@hourly', 'Catalog::update_prices')); // takes ~2 hours

Cron::set('generate_simple_html', array('* * * * *', 'Site::regenerate_simple_page')); // takes few seconds

In common case you'll get some simple html page (from Site::regenerate_simple_page() method) only after previous tasks would be complited, it's about ~4 hours. Every next Cron::run() execution will be halted while lock-file exists.

1. We can add group identifier for every job:

   Cron::set('reindex_catalog', array('@hourly', 'Catalog::regenerate_index'), 'catalog_tasks'); Cron::set('update_catalog_prices', array('@hourly', 'Catalog::update_prices'), 'catalog_tasks');

   Cron::set('generate_simple_html', array('* * * * *', 'Site::regenerate_simple_page'), 'quick_tasks');

2. And put 2 rules to crontab for every new stream

   • • • • • /usr/bin/php -f /path/to/kohana/modules/cron/run.php catalog_tasks
   • • • • • /usr/bin/php -f /path/to/kohana/modules/cron/run.php quick_tasks
   • • • • • /usr/bin/php -f /path/to/kohana/modules/cron/run.php # will handle default jobs, without group identifier

First cron process will start Cron::run() method for catalog_tasks group and it will take about 4 hours before it start next Cron::run() method for quick_tasks group

But second cron process (at next minute) will skip catalog_task (because lock-file for that group exists) and execute quick_tasks jobs

Session fix for cron

If your callbacks uses sessions (via Session class), you will always get this exception:

Error reading session data.

This is expected behavior. But its a problem for cron callbakcs execution. Solution is to overload Session::read() method in your application/classes folder:

abstract class Session extends Kohana_Session
{
    public function read($id = NULL)
    {
        try {
            parent::read($id);
        } catch (Session_Exception $e) {
            // ignore
        }
    }
}