# Runbook: [Service name] - [Failure or operation]

> Training template. Free to use, no account needed. Fill every section before
> relying on this in a real on-call. Nothing here connects to a real service.

## 1. Purpose, scope and business impact

- Purpose (one sentence):
- In scope:
- Out of scope:
- Business impact if this fails:

## 2. Symptoms and example signals

- Alert or symptom:
- Example dashboard panel:
- Example log line (synthetic):

## 3. Preconditions and authority

- Who may intervene (role):
- Required access:
- Preconditions before touching anything:

## 4. Diagnosis (expected vs observed)

| Step | Check | Expected result | Observed result |
| --- | --- | --- | --- |
| 1 | | | |
| 2 | | | |
| 3 | | | |

## 5. Decision

- Is it related to the latest change?
- Is data compatibility confirmed?
- Is approval required before intervening? From whom?

## 6. Intervention

| Step | Action | Risk | Stop condition |
| --- | --- | --- | --- |
| 1 | | | |
| 2 | | | |

## 7. Rollback (when it does NOT apply, say so)

- Rollback precondition:
- Rollback steps:
- Cases where rollback is wrong (for example irreversible data change):

## 8. Verification

- Error-rate check:
- Functional check:
- Example record check:

## 9. Stop and escalate

- Stop conditions:
- Escalate to (role):
- Communication note:

## 10. Incident record and follow-up

- Incident log link or ID:
- What to update after the event:
- Last tested date:
- Version:
- Review note: