runbook-creator

Runbook Creator

Templates and best practices for creating effective operational runbooks.

When to Use This Skill

• Creating runbooks for new services
• Documenting incident response procedures
• Writing operational playbooks
• Standardizing on-call documentation
• Automating common procedures

Runbook Principles

1. Actionable: Every step should be executable
2. Testable: Verify each step works
3. Current: Update when systems change
4. Accessible: Available during incidents (not behind VPN-only)
5. Linked: Referenced from alerts

Standard Runbook Template

Copy and customize this template:

# [Service Name] - [Issue Type]

## Overview
Brief description of what this runbook addresses.

**Last Updated**: YYYY-MM-DD
**Owner**: [Team/Person]
**Related Alerts**: [Alert names that link here]

## Symptoms
What indicates this issue is occurring:
- [ ] Symptom 1
- [ ] Symptom 2
- [ ] Symptom 3

## Impact
- **Users Affected**: [Description]
- **Severity**: [SEV1/SEV2/SEV3/SEV4]
- **Business Impact**: [Description]

## Prerequisites
- Access to [system/tool]
- Permissions: [required permissions]
- Tools: [required CLI tools]

## Diagnostic Steps

### Step 1: [Verify the Issue]
```bash
# Command to run
kubectl get pods -n production | grep -v Running

Expected Output: [What you should see] If Different: [What to do]

Step 2: [Gather Information]

# Command to run
kubectl logs deployment/my-service -n production --tail=100

Look For: [What to look for in output]

Resolution Steps

Option A: [Quick Fix - e.g., Restart]

Use when: [conditions]

# Step 1: Restart the service
kubectl rollout restart deployment/my-service -n production

# Step 2: Verify pods are coming up
kubectl get pods -n production -w

Verification: [How to confirm fix worked]

Option B: [Rollback]

Use when: [conditions]

# Step 1: Check rollout history
kubectl rollout history deployment/my-service -n production