Workers allow you to offload processes from your web service and run them in the background. They are generally used for heavy or long-running processes like mail, billing, image processing, etc. After reading this doc, you should understand:
- The basic infrastructure of workers
- How to deploy workers
- How to configure workers
Whenever code is deployed, workers are built using the same codebase as your web service. All code necessary to run your workers should be included in your codebase.
While workers are built using the same codebase as your web service(s), they have no web server or publicly accessilble end-point. Workers can only be accessed from within your app or through an SSH Connection. They're usually setup up to pull jobs from some kind of job queue. Queues can be handled in many different ways – using Redis' pub/sub functionality, storing jobs in your database or in your network filesystem, etc. It all depends on how you want to handle it.
Workers are scaled in exactly the same way as web services. You can increase the number of instances and/or the amount of RAM per instance. Which method of scaling is best depends on what your worker does. For tasks requiring a lot of RAM, you'll want to increase the amount of RAM per instance. For "light" but highly concurrent tasks, you're better off to increase your number of worker instances.
Workers allow you to run tasks in the background without timeouts. This makes them ideal for running processes that may affecting the performance of your site if run on a web service – anything from sending email, building reports, or processing images. Workers are most often accompanied by a job queue, from which they pull jobs. As mentioned above, job queues can come in many different forms.
Below are links to example PHP workers and some great tools to help with implementing PHP workers:
Deploying workers is done by including a worker service in your Boxfile.
Workers in the Boxfile YAML
worker1: type: php version: 5.5 exec: "php workers/mail.php"
All configuration for workers is handled in the Boxfile. Which config options you use depends on worker "type," or what language is powering your worker. All that's needed to get a worker up and running is a "type" and "exec," the command to start your worker process.
The exec is the command that will start your worker process.
worker1: type: php exec: "php workers/mail.php"
You Must Declare Your Executable
In your worker exec command, the executable must be declared. This can be done in one of two ways:
Include the executable at the beginning of your command.
Declare in it your script with a shebang.
Language & Service Specific Settings
Each language available in worker services has its own unique Boxfile configuration options. There are also 3rd party services with configuration options available in the Boxfile. The following docs outline each:
View the available PHP settings in the PHP Settings in the Boxfile doc.
New Relic Settings
If you have any questions, suggestions, or corrections, let us know.