Quick reference tables for common SRE concepts.
| Signal | What | How to Measure | Alert When |
|---|
| Latency | Request duration | p50, p95, p99 | p95 > SLO |
| Traffic | Requests/sec | Rate over time | Unusual spike/drop |
| Errors | Failed requests | Error rate % | > Error budget |
| Saturation | Resource usage | CPU, memory, disk | > 80% sustained |
| Level | Customer Impact | Response Time | Example |
|---|
| SEV-1 | Complete outage | Immediate | API down, database offline |
| SEV-2 | Degraded service | < 15 min | Elevated latency, partial outage |
| SEV-3 | Minor impact | < 2 hours | Single feature broken |
| SEV-4 | No impact | Business hours | Internal tool issue |