DefaultJobQueuePlugin
DefaultJobQueuePlugin
A plugin which configures Vendure to use the SQL database to persist the JobQueue jobs using the SqlJobQueueStrategy. If you add this plugin to an existing Vendure installation, you'll need to run a database migration, since this plugin will add a new "job_record" table to the database.
Example
import { DefaultJobQueuePlugin, VendureConfig } from '@vendure/core';
export const config: VendureConfig = {
// Add an instance of the plugin to the plugins array
plugins: [
DefaultJobQueuePlugin,
],
};
Configuration
It is possible to configure the behaviour of the SqlJobQueueStrategy by passing options to the static init()
function:
pollInterval
The interval in ms between polling for new jobs. The default is 200ms. Using a longer interval reduces load on the database but results in a slight delay in processing jobs. For more control, it is possible to supply a function which can specify a pollInterval based on the queue name:
Example
export const config: VendureConfig = {
plugins: [
DefaultJobQueuePlugin.init({
pollInterval: queueName => {
if (queueName === 'cart-recovery-email') {
// This queue does not need to be polled so frequently,
// so we set a longer interval in order to reduce load
// on the database.
return 10000;
}
return 200;
},
}),
],
};
concurrency
The number of jobs to process concurrently per worker. Defaults to 1.
backoffStrategy
Defines the backoff strategy used when retrying failed jobs. In other words, if a job fails and is configured to be re-tried, how long should we wait before the next attempt?
By default, a job will be retried as soon as possible, but in some cases this is not desirable. For example, a job may interact with an unreliable 3rd-party API which is sensitive to too many requests. In this case, an exponential backoff may be used which progressively increases the delay between each subsequent retry.
Example
export const config: VendureConfig = {
plugins: [
DefaultJobQueuePlugin.init({
pollInterval: 5000,
concurrency: 2
backoffStrategy: (queueName, attemptsMade, job) => {
if (queueName === 'transcode-video') {
// exponential backoff example
return (attemptsMade ** 2) * 1000;
}
// A default delay for all other queues
return 1000;
},
setRetries: (queueName, job) => {
if (queueName === 'send-email') {
// Override the default number of retries
// for the 'send-email' job because we have
// a very unreliable email service.
return 10;
}
return job.retries;
}
}),
],
};
class DefaultJobQueuePlugin {
init(options: DefaultJobQueueOptions) => Type<DefaultJobQueuePlugin>;
}
init
(options: DefaultJobQueueOptions) => Type<DefaultJobQueuePlugin>
DefaultJobQueueOptions
Configuration options for the DefaultJobQueuePlugin. These values get passed into the SqlJobQueueStrategy.
interface DefaultJobQueueOptions {
pollInterval?: number | ((queueName: string) => number);
concurrency?: number;
backoffStrategy?: BackoffStrategy;
setRetries?: (queueName: string, job: Job) => number;
useDatabaseForBuffer?: boolean;
gracefulShutdownTimeout?: number;
}
pollInterval
number | ((queueName: string) => number)
200
The interval in ms between polling the database for new jobs. If many job queues are active, the polling may cause undue load on the database, in which case this value should be increased to e.g. 1000.
concurrency
number
1
How many jobs from a given queue to process concurrently.
backoffStrategy
The strategy used to decide how long to wait before retrying a failed job.
setRetries
(queueName: string, job: Job) => number
When a job is added to the JobQueue using JobQueue.add()
, the calling
code may specify the number of retries in case of failure. This option allows
you to override that number and specify your own number of retries based on
the job being added.
Example
setRetries: (queueName, job) => {
if (queueName === 'send-email') {
// Override the default number of retries
// for the 'send-email' job because we have
// a very unreliable email service.
return 10;
}
return job.retries;
}
useDatabaseForBuffer
boolean
If set to true
, the database will be used to store buffered jobs. This is
recommended for production.
When enabled, a new JobRecordBuffer
database entity will be defined which will
require a migration when first enabling this option.
gracefulShutdownTimeout
number
20_000
The timeout in ms which the queue will use when attempting a graceful shutdown. That means when the server is shut down but a job is running, the job queue will wait for the job to complete before allowing the server to shut down. If the job does not complete within this timeout window, the job will be forced to stop and the server will shut down anyway.