Stage 6 · Operate
Alerting Hygiene & Paging Design
Handoff Readiness
Shift summaries, active incidents, risky deploys, ownership changes, and follow-the-sun transfer notes.
Why Handoff Readiness Matters
Poor handoffs are a leading cause of incident escalation. When the incoming engineer lacks context, they respond slower, make worse decisions, and miss critical information. Handoff readiness is a reliability practice.
A well-prepared handoff transfers not just status but judgment. The incoming engineer should understand not just what is happening, but why decisions were made and what to watch for.
Shift Summaries
## Shift Summary: [Date] [Shift]
### Outgoing Engineer: [name]
### Incoming Engineer: [name]
### Shift Health
- [ ] All services healthy
- [ ] No active incidents
- [ ] No risky deploys pending
- [ ] Disk space adequate
- [ ] Certificates valid > 30 days
### Events This Shift
- [HH:MM] [Event description]
- [HH:MM] [Event description]
### Active Incidents
- [Incident ID]: [Status, next steps, link]
### Risky Deploys
- [Service]: [version] deployed at [time]
- Status: [monitoring/healthy/issue]
- Rollback plan: [description]
### Watch Items
- [Item to monitor]: [reason]
### Follow-Up Actions
- [Action]: [owner, due date]
### Notes
[Any additional context for incoming engineer]Active Incident Handoff
## Active Incident Handoff: [INC-ID]
### Incident Summary
- **Title:** [Brief description]
- **Severity:** SEV[X]
- **Started:** [timestamp]
- **Duration:** [current duration]
### What Happened
[Clear description of the incident]
### What Was Done
1. [Action 1]: [result]
2. [Action 2]: [result]
### Current State
- **Error rate:** [current percentage]
- **Latency:** [current p99]
- **Impact:** [current user impact]
### Next Steps
1. [Next action to take]
2. [Alternative if first action fails]
### Key Context
- **Root cause hypothesis:** [description]
- **What NOT to try:** [failed approaches]
- **Escalation path:** [who to contact]
### Links
- Grafana: [URL]
- War room: [channel]
- PagerDuty: [URL]Risky Deploy Flags
Flag risky deploys during handoffs. A deploy that changes database schemas, modifies authentication, or touches critical paths needs extra attention from the incoming engineer.
risky_deploy_flags:
high_risk:
- "Database schema migration"
- "Authentication/authorization changes"
- "Payment processing changes"
- "Data pipeline changes"
- "Infrastructure changes (Terraform)"
- "Certificate or DNS changes"
medium_risk:
- "New feature deployment"
- "API endpoint changes"
- "Configuration changes"
- "Dependency upgrades"
low_risk:
- "Documentation updates"
- "Test changes"
- "CI/CD pipeline changes"
- "Non-production changes"
handoff_requirements:
high_risk:
- "Explicitly mention in handoff summary"
- "Provide rollback procedure"
- "Monitor for 30 minutes after deploy"
- "Have rollback command ready"
medium_risk:
- "Mention in handoff summary"
- "Provide rollback procedure"
- "Monitor for 15 minutes after deploy"
low_risk:
- "Mention in shift summary"Ownership Changes
When service ownership changes between teams, handoff documentation must include all context: known issues, architectural decisions, operational procedures, and contact information.
ownership_change_handoff:
documentation:
- "Service architecture diagram"
- "Runbook links for common operations"
- "Alert configuration and routing"
- "Known issues and workarounds"
- "Technical debt inventory"
- "Cost and capacity information"
access:
- "Kubernetes cluster access"
- "Cloud provider access"
- "Monitoring and alerting access"
- "Source code repository access"
- "CI/CD pipeline access"
- "PagerDuty service access"
contacts:
- "Original service developers"
- "Domain experts"
- "Vendor contacts"
- "Escalation paths"
verification:
- "Verify access works"
- "Verify monitoring works"
- "Verify alerts route correctly"
- "Verify runbooks are accessible"
- "Run through common operations"Follow-the-Sun Transfer Notes
## Follow-the-Sun Transfer: [Date]
### From: [Timezone] Team
### To: [Timezone] Team
### System Status
- **Overall health:** [Green/Yellow/Red]
- **Active incidents:** [count or "None"]
- **Services requiring attention:** [list]
### What Happened This Shift
1. [Event 1]: [details]
2. [Event 2]: [details]
### What Needs Attention
- [Item 1]: [specific action needed]
- [Item 2]: [specific action needed]
### What to Watch
- [Metric/trend to monitor]: [reason]
### Context for Next Shift
[Any information that helps the next shift make better decisions]
### Contacts
- **Escalation:** [name, contact]
- **Subject matter expert:** [name, contact]
### Acknowledgment
- [ ] Incoming team has reviewed this transfer
- [ ] Incoming team has questions answered
- [ ] Handoff is completeAutomate as much of the handoff as possible. Generate reports of recent deploys, system health, and active alerts. This reduces manual effort and ensures nothing is missed.
Mark this lesson complete to store local progress and unlock a cleaner resume path the next time you visit.