R-#32 Monitoring and Alerting for sivel.xyz

[vtamara]

Context

The /api/health/credentials endpoint already exists and checks Celo and Base contract reachability via nextTokenId. This issue proposes integrating it into a monitoring system and extending it.

Tasks

1. Uptime Monitoring

• Configure an external uptime monitor (e.g., UptimeRobot, BetterStack, or a cron job on adJ) to call GET /api/health/credentials every 5 minutes.
• Alert via email/Telegram if either Celo or Base contract becomes unreachable for >2 consecutive checks.
• The endpoint returns 200 if at least one contract is healthy, 503 if both are down.

2. Database Health

• Add database connectivity check to /api/health/credentials:
  • Query credential_metadata table (e.g., SELECT COUNT(*))
  • Return db: true/false in the response
• Add credential_emission table check (verify recent emissions exist)

3. RPC Health

• Add RPC latency measurement to the health check:
  • Time the nextTokenId call
  • Include celo_rpc_ms and base_rpc_ms in the response

4. Monitoring Dashboard (optional)

• Simple HTML dashboard at /health showing:
  • Green/red indicators for Celo contract, Base contract, DB, RPC latency
  • Last 10 check results
  • Auto-refresh

5. Alerting on Mint Failures

• Log mint transaction hashes to credential_emission on success
• Monitor for gaps in emission (e.g., no new emissions in last 24h for any token type)
• Alert if a mint transaction reverts (API endpoint could log failures)

6. Contract Event Monitoring

• Listen for CredentialMinted, CredentialRevoked, CredentialTypeRegistered events
• Store event logs for audit trail
• Alert on unexpected CredentialRevoked events

Acceptance Criteria

• /api/health/credentials returns: { celo, base, db, celo_rpc_ms, base_rpc_ms, timestamp }
• External monitor checks every 5 min
• Alert on 2 consecutive failures
• RPC latency tracked

[andrianvaleanu]

Setting up an external check for those endpoints is smart. Try https://pulsetic.com/ for easy stattus alerts and logs.