MBS Uptime Monitor

The public-facing side of the monitor is designed as a status center: visitors can quickly understand whether systems are healthy, which services are affected, what incidents are active, and how recent the status snapshot is. I kept the preview practical rather than decorative because this kind of tool has to be trusted under pressure.

The main repository contains the self-hosted Laravel monitor: public status pages, private admin tools, check scheduling, incident publishing, subscriber flows, API documentation, and operational docs. The package repository contains mbs047/laravel-status-probe, the lightweight package installed inside Laravel apps that need to be monitored.

After installing the package with composer require mbs047/laravel-status-probe and running php artisan status-probe:install, a Laravel app can expose bearer-protected GET /status/health and GET /status/metadata endpoints. The monitor can then pull app health, read component metadata, and map checks such as checks.db.status back into component-level status.

The monitor supports HTTP, SSL, DNS, and TCP checks. HTTP checks can validate expected status codes, latency, required text, JSON assertions, basic or bearer auth, and JSON status paths coming from package-enabled Laravel apps. Due checks are dispatched on schedule, executed through queued jobs, stored as check runs, and rolled up into component and service health.

Automated checks detect failures, but they do not automatically publish incidents. Admins can draft, publish, update, and resolve incidents through the operations panel, while subscribers receive confirmation, incident, update, and resolution emails only for published communication.

The monitor can pull metadata from a remote package-enabled app, accept a pushed registration payload from php artisan status-probe:register, or combine both approaches. That makes the package useful in different deployment contexts: the monitor can discover apps, apps can register themselves, or teams can use hybrid sync when they want stronger control.

The system treats monitoring as operational software. First-admin setup is protected by APP_SETUP_TOKEN, probe registration uses monitor-side bearer tokens, remote probe endpoints are private by default, and production docs cover cron, queue workers, mail, backups, retention, and deployment paths. The repository also includes an OpenAPI 3.1 contract and example payloads for public API consumers.

This project combines product thinking with reusable package design. The monitor gives teams a clear operational surface, while the package lets other Laravel apps participate without copying status logic into every codebase. It shows how I structure open-source work: public-facing UX, admin workflows, queue-backed jobs, package contracts, documentation, and tests all designed as one system.