SRE Thailand Bootcamp

From Ericsson Engineer to High-Paid SRE

Appendix: Ready-to-Use Incident Response Template

Copy this into INCIDENT_RESPONSE.md in your GitHub repo.

Incident Response Process

Severity Levels

SeverityDescriptionResponse TimeExamples
SEV1Complete service outage or data loss< 15 minAll users cannot place orders
SEV2Partial outage or degraded performance< 30 minOrders API returns 50x for 10% of requests
SEV3Non-critical functionality broken< 4 hrsDashboard shows incorrect data
SEV4Cosmetic or minor issueNext business dayTypo in UI text

Steps

  1. Detect — Alert fires or user reports issue
  2. Acknowledge — Respond in Slack #incident channel
  3. Triage — Determine severity, assign Incident Commander
  4. Mitigate — Stop the bleeding (rollback, feature flag, scale up)
  5. Resolve — Confirm service is healthy
  6. Postmortem — Write blameless postmortem within 48 hours

On-Call Schedule

WeekPrimarySecondary
Week APerson 1Person 2
Week BPerson 2Person 1

Escalation