Skip to content

Getting started

The common case: you want to see what your queue is actually doing, per job class, right now.

Everything the dashboard shows is collected the moment the package is installed, nothing to instrument, no jobs to change. Two steps stand between a fresh install and a working view.

Authorize yourself. The dashboard route is gated, so define the Gate in a service provider:

php
use Illuminate\Support\Facades\Gate;

Gate::define('viewQueueInsights', fn ($user) => $user->isAdmin());

Run a worker, and give it something to do. Metrics come from jobs that actually run:

bash
php artisan queue:work

Then open /queue-insights. Each job class carries its throughput, its duration spread, its failure count, and the live depth of the queue it runs on.

The number to look at first is depth against throughput. A queue whose depth climbs while throughput stays flat is not keeping up, and the job class holding the longest durations is usually why.

Next

  • Dashboard: scoping by connection, the panels, and what each metric counts.
  • Jobs, batches and chains: how the three are tracked, and what a chain reports.
  • Alerting: turn a threshold into a notification rather than watching a page.
  • Payload capture: off by default, and worth reading before turning it on.