iYup

I built iYup to show how a small monitoring stack fits together: active checks, Prometheus scraping, Grafana dashboards, Alertmanager wiring, a status API, Docker Compose, and Helm packaging.

It is not a SaaS clone. It is a small lab with real commands, real screenshots, and a verification path that can be rerun locally.

What It Proves

• active HTTP checks from ping-agent
• Prometheus metrics from ping-agent and api-gateway
• status and target APIs over live monitoring data
• latency percentiles and windowed uptime summaries
• Grafana dashboard provisioning in Docker Compose
• Alertmanager wiring from Prometheus rules
• local Docker Compose startup and Helm render checks

Run It Locally

cp .env.example .env
docker compose up -d
./scripts/verify-local.sh

Open:

• API: http://localhost:8080/status
• Prometheus: http://localhost:9090/targets
• Grafana: http://localhost:3000
• Alertmanager: http://localhost:9093

If .env overrides host ports, use those values instead. The copyable commands are in docs/DEMO_COMMANDS.md.

System Flow

flowchart TD
    A[Monitored Targets] --> B[Ping Agent]
    B --> C[Prometheus Metrics]
    C --> D[Prometheus]
    D --> E[Grafana]
    D --> F[Alert Rules]
    F --> G[Alertmanager]
    C --> H[API Gateway]
    H --> I[Status API]

Loading

Verification

Run:

./scripts/verify-local.sh

Phase 0 command logs are in docs/PHASE_0_VERIFICATION.md.

The current proof status is in docs/PORTFOLIO_PROOF_CHECKLIST.md.

Demo Screenshots

The screenshots live in docs/screenshots/. The capture commands are in docs/SCREENSHOT_GUIDE.md.

Full set:

• 01 repo overview
• 02 Docker Compose stack
• 03 API health check
• 04 API status response
• 05 Prometheus targets
• 06 Prometheus metrics query
• 07 Grafana dashboard
• 08 Alertmanager page
• 09 Helm template output
• 10 local verification script

API Surface

Endpoint	Purpose
GET /healthz	API health check
GET /targets	configured target list
GET /status	current status, latency, percentiles, and availability
GET /targets/{url}	per-target detail
GET /uptime-summary	lifetime success, failure, and availability
GET /uptime-summary-windowed?window=5m	Prometheus-backed uptime window
GET /metrics	Prometheus metrics for the API gateway

Configuration

Variable	Default	Purpose
PING_TARGET_URLS	https://google.com,https://github.com	comma-separated targets
PING_INTERVAL_SECONDS	30	check cadence
PING_CONCURRENCY	5	parallel workers
PING_RETRY_COUNT	2	retries before down
PROMETHEUS_QUERY_CACHE_SECONDS	15	cache TTL for windowed queries
API_GATEWAY_PORT	8080	API host port
PING_AGENT_PORT	18080	ping-agent host port
PROMETHEUS_PORT	9090	Prometheus host port
GRAFANA_PORT	3000	Grafana host port
ALERTMANAGER_PORT	9093	Alertmanager host port
GRAFANA_PASSWORD	admin	Grafana admin password in Docker Compose

Limitations

• no distributed tracing
• no log aggregation
• no real notification credentials
• no incident workflow tooling
• no multi-region behavior
• Helm is render-validated here, not cluster-validated

Docs

• docs/QUICKSTART.md
• docs/API.md
• docs/ARCHITECTURE.md
• docs/DEPLOYMENT.md
• docs/GRAFANA.md
• docs/OPERATIONS_RUNBOOK.md
• docs/TROUBLESHOOTING.md
• docs/PHASE_0_VERIFICATION.md
• docs/SCREENSHOT_GUIDE.md
• docs/PORTFOLIO_PROOF_CHECKLIST.md
• docs/DEMO_COMMANDS.md
• docs/RELIABILITY_SCENARIOS.md