What are SLO/SLI/SLA?

Question

Accepted Answer

**SLI** — Service Level Indicator (metric).
**SLO** — Service Level Objective (target).
**SLA** — Service Level Agreement (contract).

```yaml
# SLI: Availability
sli_availability = successful_requests / total_requests

# SLO: 99.9% availability
slo_availability = 0.999

# Error Budget
error_budget = 1 - slo = 0.001
error_budget_minutes = 0.001 * 30 * 24 * 60 = 43.2 minutes/month

# SLA: penalties for violation
sla_terms:
  - availability < 99.9%: credit 10%
  - availability < 99.0%: credit 30%
  - availability < 95.0%: credit 50%
```

```yaml
# Prometheus alert based on SLO
groups:
  - name: slo
    rules:
      - alert: ErrorBudgetBurnRate
        expr: |
          (
            1 - (sum(rate(http_requests_total{status=~"2.."}[1h])) /
                 sum(rate(http_requests_total[1h])))
          ) > (14.4 * 0.001)
        for: 1h
        labels:
          severity: critical
```

**Formula:** Error Budget = 1 - SLO.
If SLO = 99.9%, then Error Budget = 0.1% = 43 min/month.

DevOps: What are SLO/SLI/SLA?