Laravel Schedule Monitoring Tools Compared: Spatie, Oh Dear, Cronitor, Sentry
Contents
Laravel is the best-served framework for scheduler monitoring: several packages read Schedule::events() and turn every task into a monitor, so you don't keep a second list by hand. The real differences are where the expectations live (your own database or an outside service), what has to run at deploy time, and whether anything alerts when schedule:run stops being called at all. Only an outside service can tell you that, because when the scheduler stops, nothing inside Laravel runs to notice. This comparison covers Laravel 10 through 13. Package and vendor facts were checked against their documentation on November 16, 2026.
Built-In Hooks Come First#
Before choosing a package, know what the framework already does. Each scheduled task can ping a URL and run callbacks (Laravel task scheduling):
use Illuminate\Support\Facades\Schedule;
Schedule::command('reports:generate')
->dailyAt('02:00')
->pingBefore('https://heartbeat.example.com/reports/start')
->pingOnSuccess('https://heartbeat.example.com/reports/success')
->pingOnFailure('https://heartbeat.example.com/reports/fail');That is enough for a few tasks and any heartbeat service. It doesn't scale well: each URL is written by hand, each check is created on the service by hand, and a task removed from routes/console.php leaves a check that alerts until someone deletes it. How to monitor Laravel scheduled tasks covers the hooks, onFailure, and output capture in detail.
spatie/laravel-schedule-monitor#
Spatie's package is the most widely installed option, at about 450,000 Packagist downloads a month when checked (repository).
composer require spatie/laravel-schedule-monitor
php artisan schedule-monitor:sync
php artisan schedule-monitor:list- Discovery.
schedule-monitor:syncreads your schedule and stores each task in amonitored_scheduled_taskstable; run it on every deploy. Starts, finishes, failures, and skips are logged tomonitored_scheduled_task_log_items. - Naming. Commands use their command name and queued jobs their class; closures need
->monitorName('...').->doNotMonitor()excludes a task. - Late detection. A task that hasn't finished by its scheduled time plus a grace time (5 minutes by default,
->graceTimeInMinutes()per task) shows as late inschedule-monitor:list. - Alerting. The package records state in your own database. Notifications for a task that didn't run come from Oh Dear, which receives the pings when you set
api_tokenandmonitor_idin the package config.
On its own, the package is a local record you can inspect, and its lateness check runs inside the same application that might be down. With Oh Dear connected, the expectations live outside and alert.
Oh Dear#
Oh Dear is the hosted side of Spatie's package, part of a wider monitoring product that also covers uptime, certificates, and status pages. Its cron monitoring accepts start, finish, and failure pings with runtime, memory, and exit code, and schedules as a frequency or a cron expression with a time zone (Oh Dear cron monitoring). Pricing is per site, starting at $17 a month for 2 sites with 5 cron jobs per site, and there is no free plan (pricing).
One behavior to know: schedule-monitor:sync removes Oh Dear cron monitors that are no longer in your schedule, unless you pass --keep-old. That keeps the list in step with the code, and it also means a task deleted by mistake loses its monitor on the next deploy without an alert.
cronitor/laravel#
Cronitor's Laravel package works the same way against Cronitor's service (repository):
composer require cronitor/laravel
php artisan vendor:publish --tag=cronitor-config
php artisan cronitor:sync --dry-run
php artisan cronitor:synccronitor:sync builds monitors from Schedule::events() and is meant to run at deploy time, before a task's first run. Scheduled tasks send run, complete, and fail events, and queue jobs send the same. ignored_tasks and ignored_jobs in config/cronitor.php exclude tasks, and CRONITOR_API_KEY is required. Cronitor's free plan covers 5 monitors; paid monitors are priced per monitor plus per user. See the Cronitor comparison.
Sentry Crons for Laravel#
Sentry's Laravel SDK adds a sentryMonitor() macro to scheduled tasks (Laravel Crons):
use Illuminate\Support\Facades\Schedule;
Schedule::command('emails:send')
->everyHour()
->sentryMonitor(
checkInMargin: 5,
maxRuntime: 15,
);It's opt-in per task: tasks without the macro aren't monitored. By default the SDK infers the schedule and sends it with each check-in; updateMonitorConfig: false makes the Sentry UI the only source, and then a monitorSlug is required. Tasks using between, unlessBetween, when, or skip aren't supported. Every Sentry plan includes one cron monitor, and more cost $0.78 each from a paid plan's budget. See the Sentry Crons comparison.
CronRadar#
CronRadar is a monitor for scheduled jobs only. The cronradar/laravel package (0.1.3, Laravel 10 through 13) reads the schedule at boot, declares every task with its cron expression, and records start, completion, and failure through the schedule's before, after, and onFailure hooks (Laravel integration guide):
composer require cronradar/laravelCRONRADAR_API_KEY=ck_app_xxxxxxxxxxxxxxxxxxxx// routes/console.php
use Illuminate\Support\Facades\Schedule;
Schedule::command('reports:generate')->dailyAt('02:00');
Schedule::call(fn () => app(\App\Services\Billing::class)->closeDay())
->dailyAt('23:55')
->monitor('close-billing-day');
Schedule::command('internal:cleanup')
->daily()
->skipMonitor();- Every task by default. Commands, closures, queued jobs, and
exectasks are all discovered;->skipMonitor()opts one out and->monitor('key')names it. Closure keys otherwise come from the file and line, so name the ones you care about. - No deploy step required. The package syncs when Laravel boots.
- Never-ran tasks. Monitors are declared before the first run, so a task that never runs alerts after its first expected time plus grace.
- Scheduler stopped. When cron stops calling
schedule:run, or theschedule:workcontainer dies, no starts arrive and every task's monitor goeswarningthencritical. - Time zones. CronRadar evaluates schedules in the CronRadar application's IANA time zone, UTC if unset. Set it to match
->timezone()or your app's schedule time zone. - Removed tasks. The README says removed tasks stay in CronRadar until you delete them. Delete or pause them when you drop a task, or they alert as missed.
Alerts go to email, Slack, webhook, Discord, or PagerDuty. See the Laravel page, the PHP overview, and pricing.
Side by Side#
| Option | Discovery | Deploy step | Where expectations live | Alerts when scheduler stops | Free plan |
|---|---|---|---|---|---|
| Built-in ping hooks | Per task, by hand | No | Heartbeat service | Yes, per configured check | Depends on service |
| spatie/laravel-schedule-monitor | schedule-monitor:sync |
Yes | Your database | Not on its own | Open source |
| Spatie + Oh Dear | schedule-monitor:sync |
Yes | Oh Dear | Yes | No |
| cronitor/laravel | cronitor:sync |
Yes | Cronitor | Yes | 5 monitors |
Sentry sentryMonitor() |
Per task | No | Sentry | Yes, for tasks with the macro | 1 monitor |
| cronradar/laravel | At boot | No | CronRadar | Yes | See pricing |
Which to Choose#
- You already pay for Oh Dear: Spatie's package plus Oh Dear is the established path, and it bundles uptime and certificate checks if you want them.
- You want a record in your own database: Spatie's package alone, with the understanding that it can't alert about its own application being down.
- You're on Sentry with a few critical tasks:
sentryMonitor()on those tasks. - You want every task covered without a deploy step, separate from uptime tooling: cronradar/laravel.
Whatever you choose, confirm the scheduler itself is triggered; Laravel schedule not running lists the eight ways it isn't.
Frequently Asked Questions#
How do I get alerted when a Laravel scheduled task doesn't run?#
Use a monitor outside the application that knows each task's schedule. Laravel's pingBefore and pingOnSuccess hooks send heartbeats to any service, and packages such as cronitor/laravel, Spatie with Oh Dear, and cronradar/laravel create one monitor per task from the schedule. An in-app check can't alert when the scheduler or the server stops.
Does spatie/laravel-schedule-monitor send notifications?#
It records task runs in your database and shows late or failed tasks with schedule-monitor:list. Notifications come through its Oh Dear integration, which receives the pings and alerts when a task is late or fails.
Do I need to run a sync command on every deploy?#
For Spatie's package and cronitor/laravel, yes: their sync commands build the monitors from the current schedule, and both docs say to run them at deploy time. Sentry sends configuration with check-ins instead, and cronradar/laravel syncs when Laravel boots.
What happens to monitors when I remove a scheduled task?#
It depends on the tool. Spatie's sync removes Oh Dear monitors that are no longer scheduled unless you pass --keep-old. With cronradar/laravel 0.1.3, removed tasks stay until you delete them. With manual hooks, the heartbeat check keeps expecting pings until you remove it.
Put this guide to work
Connect your Laravel job to monitoring
Follow the current integration guide, check the release limitations, and verify a real run and alert delivery.
Packagist installs cronradar/laravel 0.1.3 for Laravel 10 through 13, which discovers scheduled tasks.