Laravel Horizon must use Redis as queue driver, after installation, configure through php artisan horizon:install, set QUEUE_CONNECTION=redis, start php artisan horizon after publishing resources, access dashboard through /horizon, 1. Use Gate to control access in HorizonServiceProvider, 2. View recent tasks through "Recent Jobs", 3. Retry or delete failed tasks in "Failed Jobs", 4. Use "Metrics" to analyze processing rates and bottlenecks, 5. Monitor specific task classes in "Monitoring", 6. Manage worker status through "Workers", 7. Configure multiple supervisors and queue balancing policies in config/horizon.php, 8. Set up Slack or mail notifications to receive failures or performance alerts, 9. Add tags to tasks for tracking, 10. Restart Horizon regularly with php artisan horizon:terminate to ensure Redis memory is properly managed for comprehensive queue monitoring.

Laravel Horizon is a powerful dashboard and configuration tool for managing and monitoring Laravel's Redis-based queues. It gives you deep insights into job throughput, runtime, failed jobs, and worker performance—all through a clean, real-time UI. Here's how to set it up and use it effectively for monitoring your queues.

? Install and Configure Horizon

First, make sure you're using Redis as your queue driver since Horizon only works with Redis.

1. Install Horizon via Composer:

   

    composer requires laravel/horizon

2. Publish the Horizon configuration and assets:

    php artisan horizon:install
   php artisan vendor:publish --provider="Laravel\Horizon\HorizonServiceProvider"

   This creates a config/horizon.php file.

   

3. Set your queue driver to redis in .env :

    QUEUE_CONNECTION=redis

4. Ensure Redis is configured properly in config/database.php .

?? Start Horizon and Access the Dashboard

1. Start the Horizon process:

    php artisan horizon

   This starts the Horizon queue worker (replace php artisan queue:work in your production processes).

2. Access the dashboard: By default, Horizon is available at:

    http://your-app.test/horizon

   ? Access is restricted. You can customize who can view the dashboard in app/Providers/HorizonServiceProvider.php using gate logic.

   Example (allow only admins):

    Gate::define('viewHorizon', function ($user) {
       return in_array($user->email, [
           'admin@yoursite.com'
       ]);
   });

? Monitor Queues Using the Horizon Dashboard

Once logged in, you'll see several key tabs:

1. Recent Jobs

• Shows jobs processed in the last 5 minutes.
• Includes job name, runtime, status (completed/failed), and timestamp.
• Helps spot slow or failing jobs quickly.

2. Failed Jobs

• Lists all failed jobs with full exception trace.
• You can:
  • Retry individual jobs.
  • Delete them.
  • Automatically notify via email or Slack (if configured).

3. Metrics

• Visual graphs showing:
  • Jobs processed per minute.
  • Job run time averages.
  • Queue wait times.
• Great for identifying bottlenecks over time.

4. Monitoring

• Allows you to "watch" specific job classes.
• Shows real-time stats: total processed, failed, runtime.
• Useful for high-priority jobs like email sends or webhooks.

5. Workers

• Shows active Horizon workers.
• Displays CPU, memory, and job counts.
• You can pause/resume queue processing from here.

6. Master

• View all Horizon processes (supervisors and workers).
• Useful for debugging scaling or deployment issues.

?? Configure Horizon for Better Monitoring

In config/horizon.php , you can define supervisor and queue behavior:

 'environments' => [
    'production' => [
        'supervisor-1' => [
            'connection' => 'redis',
            'queue' => ['default'],
            'balance' => 'auto',
            'processes' => 10,
            'tries' => 3,
        ],
    ],
],

• Use multiple supervisors for different queues (eg, high , default , low ).
• Set balance to auto , simple , or false depending on load.
• Horizon automatically adjusts process count based on queue activity when using auto .

? Set Up Notifications

Horizon can alert you when:

• A job fails.
• A long-running job exceeds a threshold.
• Horizon is down.

Configure notifications in config/horizon.php under notifications :

 'notifications' => [
    'slack' => [
        'webhook_url' => env('SLACK_WEBHOOK_URL'),
    ],
    'email' => [
        'address' => 'admin@yoursite.com',
    ],
],

Enable specific alerts:

 'failed' => true,
'metrics' => true,
'trim' => true,

Make sure your app has the Notifiable user and proper mail/slack drivers set up.

? Tips for Effective Monitoring

• Tag your jobs so they appear clearly in Horizon:

   public function tags()
  {
      return ['mail', 'user:' . $this->user->id];
  }

• Use different queues for different priorities and monitor them separately.

• Schedule regular Horizon restarts in production (eg, after deployments) using:

   php artisan horizon:terminate

  This gracefully restarts Horizon.

• Monitor Redis memory usage — Horizon stores metrics, so large apps may need Redis cleanup strategies.

---

Horizon makes Laravel queue monitoring visual, actionable, and reliable. Once set up, it's like having a control panel for your background jobs — you'll catch issues before users do.

Basically: install, configure Redis, run horizon , secure the dashboard, and start watching your queues.

The above is the detailed content of How to use Laravel Horizon for monitoring queues?. For more information, please follow other related articles on the PHP Chinese website!