监控与告警
incident-runbook
更新于 2026-09-15
Incident Runbook 工程规则
Runbook rules: detection, mitigation and evidence capture per incident class
- 技术栈
- Incident Runbook
- 主题
- 工程基线
- 条目
- 9 条
- 版本
- v0.2.0
✓ 应当这么做
- ✓ Emit structured logs with correlation ids connecting request, job and outbound call
- ✓ Use histogram metrics for latency and counters for outcomes
- ✓ Alert on symptoms (error rate, saturation) rather than causes
✕ 常见反模式
- ✕ Logging secrets, tokens or full personal payloads
- ✕ Cardinality bombs in metric labels (user ids, raw paths)
- ✕ Dashboards without an owner or documented decision threshold
◎ 交付前验证
- ◎ The new code path is visible in logs end-to-end
- ◎ Dashboard panel added or updated to match the new metric
- ◎ Alert routing tested once end-to-end