monitoring

CheesePay Uptime Monitoring

External uptime monitoring for api.cheesepay.xyz and Stellar Horizon connectivity. Alerts fire within 2 minutes of downtime; on-call is notified within 5 minutes.

---

Architecture

Internet
  │
  ├── UptimeRobot (primary, 1-min polling)
  │     ├── Monitor: GET https://api.cheesepay.xyz/health          → email + Slack
  │     ├── Monitor: GET https://api.cheesepay.xyz/health/ready    → email + Slack
  │     └── Status page: status.cheesepay.xyz (public)
  │
  └── Better Uptime (secondary / on-call escalation, 30-sec polling)
        ├── Monitor: GET https://api.cheesepay.xyz/health
        ├── Monitor: GET https://api.cheesepay.xyz/health/ready    (checks Stellar Horizon)
        └── On-call rotation → PagerDuty / phone call within 5 min

The /health endpoint returns { "status": "ok" } (always 200 while the process is alive). The /health/ready endpoint checks PostgreSQL + Stellar Horizon and returns 503 if either is down — this is the critical monitor.

---

Monitors to Configure

Monitor	URL	Method	Expected Status	Interval	Alert
API Liveness	https://api.cheesepay.xyz/health	GET	200	1 min	Email + Slack
API Readiness (DB + Stellar)	https://api.cheesepay.xyz/health/ready	GET	200	1 min	Email + Slack + PagerDuty
Stellar Horizon	https://horizon.stellar.org/	GET	200	5 min	Slack

---

Setup Guides

• UptimeRobot Setup
• Better Uptime Setup
• Status Page

---

Acceptance Criteria Verification

Criterion	How it's met
Downtime detected within 2 min	UptimeRobot 1-min interval + Better Uptime 30-sec interval
On-call alerted within 5 min	Better Uptime phone/SMS escalation after 1 failed check
Status page shows real-time component status	status.cheesepay.xyz polls /health/admin every 60s