Health Checks
The application uses spatie/laravel-health for monitoring system health. In production, the scheduler runs the checks every minute (RunHealthChecksCommand) and dispatches the queue heartbeat job every five minutes (DispatchQueueCheckJobsCommand). Neither is scheduled outside production, so run php artisan health:check when you need results there. The schedule heartbeat (ScheduleCheckHeartbeatCommand) runs every minute in every environment.
Registered Health Checks
Section titled “Registered Health Checks”The following health checks are registered by default in HealthServiceProvider:
| Check | Description | Environment |
|---|---|---|
| Database | Verifies database connectivity | Production only |
| Queue | Fails when the heartbeat job hasn’t been processed within 15 minutes | Production only |
| Cache | Verifies cache system is operational | All |
| Schedule | Verifies scheduled tasks have run recently | All |
| Debug Mode | Ensures debug mode is disabled | Non-local |
| Optimized App | Ensures app is optimized | Non-local |
| Security Advisories | Checks for known package vulnerabilities | All |
| Redis | Verifies Redis connectivity | When Redis driver is in use |
| Directory Search | Verifies Northwestern Directory Search API | All, but skipped locally when DIRECTORY_SEARCH_API_KEY is blank |
Additional checks are available for services like Redis, Horizon, Octane, and more. See the available checks documentation for a complete list. To add a check, register it in the HealthServiceProvider.
API Endpoint
Section titled “API Endpoint”A health check endpoint is available for external monitoring systems, protected by a secret token.
Endpoint: GET /api/health
Authentication: Requires an X-Secret-Token header matching the HEALTH_SECRET_TOKEN environment variable, compared in constant time by Chassis’s RequireSecretToken middleware (RequireSecretToken::class . ':health.secret_token' in routes/api.php). Until HEALTH_SECRET_TOKEN is set, the endpoint refuses every request with 403.
Freshness: Every request runs the checks and stores the results before responding, because oh_dear_endpoint.always_send_fresh_results is true in config/health.php.
curl -H "X-Secret-Token: your-secret-token" https://your-app.example.com/api/healthResponse:
{ "finishedAt": 1703299200, "checkResults": [ { "name": "database", "label": "Database", "notificationMessage": "Database is operational", "shortSummary": "OK", "status": "ok", "meta": {} } ]}HTTP Status Codes:
| Code | Meaning |
|---|---|
200 | Every check returned ok or skipped |
503 | At least one check returned warning, failed or crashed |
403 | Missing or incorrect token, or HEALTH_SECRET_TOKEN is not set |
ok and skipped count as passing; warning, failed and crashed return 503. A check skipped because it doesn’t apply in an environment (the database and queue checks outside production, Redis when the app doesn’t use it) doesn’t fail the endpoint: config/health.php sets treat_skipped_as_failure to false.
Check Status Values:
| Status | Description |
|---|---|
ok | Check passed |
warning | Check passed with warnings |
failed | Check failed |
crashed | Check threw an exception |
skipped | Check was conditionally skipped |
Notifications
Section titled “Notifications”In production, health checks run every minute through the scheduler, but email notifications are disabled by default. To receive alerts when checks fail, enable notifications and configure a recipient email address.
Enabling Notifications
Section titled “Enabling Notifications”Add the following to your .env file:
HEALTH_NOTIFICATIONS_ENABLED=trueHEALTH_NOTIFICATION_EMAIL=your-team@northwestern.eduWhile checks keep failing, a notification goes out at most once an hour (notifications.throttle_notifications_for_minutes in config/health.php).
Result Storage
Section titled “Result Storage”The application uses EloquentHealthResultStore to persist health check results in the database. This enables:
- Viewing health status in the admin panel’s Platform Overview page
- Historical tracking for trend analysis
- Results available immediately after scheduler runs
A daily model:prune for HealthCheckResultHistoryItem in routes/console.php deletes results older than keep_history_for_days (5, in config/health.php).
Adding Custom Health Checks
Section titled “Adding Custom Health Checks”Creating a Check
Section titled “Creating a Check”Create a new check class extending Spatie\Health\Checks\Check:
<?php
namespace App\Domains\Core\Health;
use Spatie\Health\Checks\Check;use Spatie\Health\Checks\Result;
class CustomServiceCheck extends Check{ protected ?string $label = 'Custom Service';
public function run(): Result { $result = Result::make();
try { $isHealthy = $this->checkService();
if ($isHealthy) { return $result ->ok() ->shortSummary('Service operational'); }
return $result ->failed('Service unavailable') ->shortSummary('Service down'); } catch (\Exception $e) { return $result ->failed($e->getMessage()) ->shortSummary('Check failed'); } }
private function checkService(): bool { // Your health check logic here return true; }}Registering the Check
Section titled “Registering the Check”Add your check to HealthServiceProvider:
Health::checks([ // ... existing checks CustomServiceCheck::new(),]);Conditional Checks
Section titled “Conditional Checks”Run checks only in specific environments:
CustomServiceCheck::new() ->if(App::isProduction()),Or skip checks in certain environments:
CustomServiceCheck::new() ->unless(App::isLocal()),