Database Postmortem Template

← DBRE Home | ← Main

Use for: operational failures — outage, failover, replication break, backup failure. Blameless. Use roles not names. → Handbook

---

Field	Value
ID	PM-YYYY-NNN
Date	YYYY-MM-DD
Severity	SEV1 / SEV2 / SEV3
DB / Cluster
Owner
Status	Draft / In Review / Approved

---

Summary

What failed, duration, how detected, how resolved. 3–5 sentences.

---

Timeline

Time (UTC)	Event
	Leadup (change, job, config that created the condition)
	Fault began
	Detected
	Mitigation started
	Service restored

Detection gap: ___    Response gap: ___

---

Five Whys

1. Why did users/services experience impact? →
2. Why? →
3. Why? →
4. Why? →
5. Why? →

Root cause:

Category: Bug / Change / Scale / Architecture / Dependency / Unknown DB sub-type: Connection handling / Replication / Mount/storage / Migration / Backup / Config drift / Monitoring gap

---

Impact

Duration	Systems affected	Users affected	Data loss	Replication impacted
			Yes / No	Yes / No

---

Corrective Actions

Actionable (verb + outcome) · Specific · Bounded (definition of done). At least one must address root cause.

#	Category	Action	Owner	Due	Ticket
1	Prevent
2	Detect
3	Mitigate

Investigate · Mitigate this incident · Repair damage · Detect future · Mitigate future · Prevent future

---

Lessons Learned

Went well	Could have gone better	Got lucky

---

Approval

Approver (role)	Root cause agreed	Actions agreed	Date
	☐	☐