API Monitoring

Sysmos runs API health checks from distributed workers across multiple global regions. This gives you a true picture of how your APIs perform for users worldwide.

How It Works

Workers (multi-region) → Redis Streams → Ingest Service → PostgreSQL + ClickHouse
                                                        ↓
                                              Notification Service
                                       (Slack, Discord, Email, Webhook, PagerDuty)

1. Workers run in multiple regions and execute HTTP checks against your endpoints
2. Results are published to Redis Streams for reliable message delivery
3. The Ingest Service consumes results, evaluates alert conditions, and stores data in PostgreSQL (configuration) and ClickHouse (time-series metrics)
4. The Notification Service sends alerts through your configured channels with cooldown and deduplication

Metrics Collected

For each check, Sysmos records:

Check Types

• HTTP/HTTPS — GET, POST, PUT, DELETE with custom headers and body
• Status code validation — Alert if response code doesn't match expected
• Response body validation — Check if response contains (or doesn't contain) specific text
• Response time threshold — Alert when latency exceeds your threshold
• SSL certificate monitoring — Alerts before certificate expiration

Data Retention

• PostgreSQL — Monitor configuration, alert rules, and recent status
• ClickHouse — Full time-series metrics for historical analysis and fast aggregation