StatusFlare: Forging a Serverless Watchtower

Introduction

Running your own infrastructure means living with the constant fear of the “Red Light” — that moment when a service goes down and you’re the last to know. StatusFlare is my solution: a serverless status dashboard that watches my services so I don’t have to.

Why StatusFlare?

Most status pages are either:

• Over-engineered: Complex setups requiring dedicated infrastructure
• Under-powered: Simple uptime checks with no real insights
• Corporate: Tied to specific monitoring services with vendor lock-in

StatusFlare aims to be different: serverless, fast, and completely under my control.

Architecture Overview

Serverless Core

• Built on Cloudflare Workers for global edge deployment
• Uses Cloudflare D1 for persistent data storage
• Cron-based monitoring every 60 seconds from multiple global locations

Monitoring Capabilities

• HTTP/HTTPS endpoint checks with customizable timeouts
• OAuth2 token support for protected APIs
• JSON response parsing for detailed error information
• Latency tracking and trend analysis

User Interface

• Glassmorphism design with Catppuccin color palette
• Real-time status updates
• Service-specific history pages
• Dynamic SVG status badges for external use

Recent Enhancements

The project has evolved significantly:

Incident Management

• Manual incident reporting system
• Link incidents to specific services
• Track resolution status and timeline

Authentication & Security

• Authelia OIDC integration for admin access
• Role-based permissions
• Secure API endpoints

Advanced Monitoring

• Custom HTTP methods and headers
• OAuth2 token acquisition and caching
• Smart response parsing for better error visibility

Developer Experience

• Real-time status badges for README files
• Detailed service history with latency trends
• Incident timeline visualization

Deployment Pipeline

StatusFlare uses a fully automated CI/CD pipeline:

• Code Changes: Push to main branch triggers pipeline
• Validation: TypeScript type-checking and linting
• Build: Compile and package for Cloudflare Workers
• Deploy: Automatic deployment via Wrangler
• Migrate: Database schema updates handled automatically

The entire process from commit to production takes under 60 seconds.

Monitoring Strategy

StatusFlare monitors my entire infrastructure:

• Core Services: Nextcloud, Gitea, Jellyfin
• Network Services: DNS, VPN endpoints
• Custom Applications: Personal projects and APIs
• External Dependencies: Third-party services I rely on

Each service gets:

• Health check endpoint
• Latency tracking
• Incident logging
• Historical performance data

Lessons Learned

Building a monitoring system taught me:

Serverless Isn’t Magic

• Cold starts can affect monitoring accuracy
• Database query limits require careful design
• Edge deployment means debugging from afar

Monitoring is Hard

• False positives are worse than false negatives
• Network issues can masquerade as service failures
• Historical data is more valuable than real-time alerts

Self-Hosting Reality

• You are your own DevOps team
• Backup plans are essential
• Documentation saves hours of debugging

Future Plans

• Alerting System: Email, SMS, and webhook notifications
• Performance Analytics: Deeper insights into service behavior
• Multi-User Support: Team collaboration features
• Plugin System: Extensible monitoring capabilities

Philosophy

StatusFlare embodies my belief that monitoring shouldn’t be a burden. It should be:

• Reliable: Always working, even when other services fail
• Informative: Providing actionable insights, not just status lights
• Accessible: Easy to understand for both technical and non-technical users
• Independent: Not dependent on the services it monitors

The goal isn’t just to know when things break, but to understand how they work and predict when they might fail.