Back to Index

Queueable Jobs

FEAST allows easily building queueable jobs to run at a later time. To create a queueable job, create a class at Jobs\Queueable that extends from Feast\Jobs\QueueableJob. Jobs by default are placed in the default queue and will try to run a maximum of 3 tries before failing. You can either override the properties directly in your inherited class or use the methods on the parent class to change them.

Methods

There are several methods for working with job parameters.

1. setQueueName - This method sets the queuename to run on.
2. setMaxTries - This method sets the maximum tries before the job is marked as failed.
3. setJobName - This method sets the job name.
4. getQueueName - This method gets the queuename to run on.
5. getMaxTries - This method gets the maximum tries before the job is marked as failed.
6. getJobName - This method gets the job name.

Storing a Job to run later

The store method inserts the job into the Jobs table on the database. All data needed to run the job MUST be part of your job class.

Running a Job

The run method on a job class will run the job. Return true on success. If the job throws an exception or the run method returns false, the tries will be incremented, and the job will be retried if the max count is not exceeded.

Listening for Jobs to run

FEAST includes a Job listener that is designed to be always on. On Linux, it is recommended to use Supervisor to manage this process. See feast:job:listen for details.

Running a single Job

Sometimes, a job fails repeatedly or gets stuck. In that situation, FEAST includes a separate command line tool to run a job directly. To do so, first check the jobs table for the job_id. Then, run the job using feast:job:run-one