Operate

Outcome: Reliable production service — observable, alertable, with defined incident response and SLOs.

Goals

• Metrics, logs, traces (OpenTelemetry-aligned)
• Golden dashboards + SLO/error budgets
• On-call with severity model and runbooks
• GuardDuty / access anomalies monitored

Key roles

Role	Accountable for
SRE	SLOs, on-call, incident commander
SEC	Security incidents, breach playbooks
DEV	Fix forward during incidents

→ DevOps perspective · Security perspective

Decision guides

• Monitoring, tracing & logging
• Dashboards & reporting
• Incident management

Topic deep dives

• Operations & observability

Reference SOPs

SOP	Procedure
SOP-007	Incident response

Pitfalls

Pitfall	Mitigation
Alerts without runbooks	Every page links to runbook in repo
PII in logs/traces	Redaction standards (Data governance)
Hero culture on incidents	Blameless postmortem → Learn

---

← Release · Lifecycle overview · Learn →