Incident Hotfix

Structured incident response, hotfix, and postmortem process.

Severity Levels

Level	Impact	Response Time	Examples
P0	System down	Immediate	Complete outage, data loss
P1	Major feature broken	< 1 hour	Auth broken, payments failing
P2	Feature degraded	< 4 hours	Slow performance, partial outage
P3	Minor issue	< 24 hours	UI bug, non-critical error

Incident Response

1. Assess

• Identify the symptom
• Determine severity level
• Check monitoring dashboards
• Review recent deployments
• Communicate status

2. Isolate

• Identify affected components
• Check error logs
• Review recent changes
• Determine blast radius

3. Mitigate

Choose one:

Rollback (safest)

• Revert to last known good state
• Apply when cause is unclear

Hotfix (targeted)

• Minimal change to fix issue
• Apply when cause is clear and fix is simple

Feature Flag (quick)

• Disable problematic feature
• Apply when feature is isolatable

4. Verify

• Issue resolved
• Error rates normalized
• Performance restored
• No side effects

5. Communicate

• Update status page
• Notify stakeholders
• Document timeline

Hotfix Process

1. Create Hotfix Branch

# From production/main branch
git checkout main
git pull origin main
git checkout -b hotfix/issue-description

2. Minimal Fix

Rules for hotfix code:

• Smallest possible change
• No refactoring
• No unrelated changes
• Must pass tests

3. Verify

bun run lint:fix
bun run build
bun run test

# Test the specific fix
# Verify in staging if time permits

4. Deploy

# Merge to main
git checkout main
git merge hotfix/issue-description

# Tag and deploy
git tag -a v1.2.4 -m "Hotfix: issue description"
git push origin main --tags

5. Backport

# Merge hotfix to develop branch
git checkout develop
git merge hotfix/issue-description
git push origin develop

Postmortem Template

Write a postmortem within 48 hours of resolution.

# Incident Postmortem: [Title]

## Summary

| Field | Value |
|-------|-------|
| Date | YYYY-MM-DD |
| Duration | X hours Y minutes |
| Severity | P0/P1/P2/P3 |
| Author | [Name] |

## Impact

- [Number of users affected]
- [Revenue impact if applicable]
- [Other business impact]

## Timeline (UTC)

| Time | Event |
|------|-------|
| HH:MM | Issue first detected |
| HH:MM | Team alerted |
| HH:MM | Root cause identified |
| HH:MM | Fix deployed |
| HH:MM | Issue resolved |

## Root Cause

[Clear, technical explanation of what caused the incident]

## Detection

How was the incident detected?
- [ ] Monitoring alert
- [ ] Customer report
- [ ] Internal discovery

Could we have detected it earlier?
- [Analysis]

## Resolution

What fixed the issue?
- [Description of fix]

Was it a rollback or hotfix?
- [Details]

## Lessons Learned

### What Went Well
- [Point 1]
- [Point 2]

### What Went Wrong
- [Point 1]
- [Point 2]

### Where We Got Lucky
- [Point 1]

## Action Items

| Action | Owner | Due Date | Status |
|--------|-------|----------|--------|
| [Action 1] | [Name] | YYYY-MM-DD | [ ] |
| [Action 2] | [Name] | YYYY-MM-DD | [ ] |

## Prevention

How do we prevent this class of issue?

- [ ] Add monitoring for [X]
- [ ] Add test for [Y]
- [ ] Improve process for [Z]

Communication Templates

Initial Alert

🚨 [P1] Investigating issues with [service]

Impact: [Brief description]
Status: Investigating
ETA: Unknown

Updates to follow.

Update

🔄 [P1] Update on [service] issue

Status: Root cause identified, fix in progress
Impact: [Updated impact]
ETA: [Time estimate]

Next update in [X] minutes.

Resolution

✅ [P1] Resolved: [service] issue

Duration: [X hours Y minutes]
Resolution: [Brief description]

Postmortem to follow within 48 hours.