Support and Incident Response
Executive Summary
RHEA operational support follows a tiered model with clear escalation paths, from content author self-service through platform-level Adobe support. Monitoring is handled via Adobe RUM analytics and custom data layer event tracking.
Support Tiers
| Tier | Scope | Team | Response |
|---|---|---|---|
| Tier 0 | Self-service | Content authors | Sidekick documentation, block reference guides |
| Tier 1 | Content and configuration issues | Site team | Content fixes, metadata adjustments, fragment updates |
| Tier 2 | Block code and styling issues | RHEA development team | Block bug fixes, CSS corrections, component model changes |
| Tier 3 | Platform and infrastructure | Adobe support / DevOps | EDS platform issues, AEM Cloud Service, CDN, DNS |
Escalation Path
| Condition | Escalate To | Channel |
|---|---|---|
| Content not rendering correctly | Tier 1 — Site team | Slack / Teams channel |
| Block behavior bug or styling defect | Tier 2 — RHEA dev team | GitHub Issue on abbvie-migrated/rhea |
| Preview/live URL not responding | Tier 3 — DevOps | Incident channel + Adobe support |
| AEM Author unavailable | Tier 3 — Adobe support | Adobe support portal |
| Performance degradation (Lighthouse scores) | Tier 2 — RHEA dev team | GitHub Issue |
| Security vulnerability in dependency | Tier 2 — RHEA dev team | GitHub Security Advisory |
Incident Response
Response Procedure
| Step | Action | Owner |
|---|---|---|
| 1 | Detect — Issue identified via monitoring, user report, or automated alert | Any team member |
| 2 | Triage — Classify severity and assign to correct tier | On-call lead |
| 3 | Investigate — Determine root cause (content, code, or infrastructure) | Assigned tier |
| 4 | Resolve — Apply fix: content unpublish/republish, code revert, or platform action | Assigned tier |
| 5 | Verify — Confirm resolution on preview and live environments | QA / Reporter |
Severity Classification
| Severity | Definition | Response Time | Resolution Target |
|---|---|---|---|
| P1 - Critical | Live site down, all users affected | Immediate | 1 hour |
| P2 - High | Major feature broken, significant user impact | 1 hour | 4 hours |
| P3 - Medium | Minor feature issue, workaround available | 4 hours | 1 business day |
| P4 - Low | Cosmetic issue, no functional impact | 1 business day | Next sprint |
Monitoring
Adobe RUM (Real User Monitoring)
| Metric | Description | Endpoint |
|---|---|---|
| Core Web Vitals | LCP, FID, CLS measurements | https://ot.aem.live |
| Page views | Page load tracking | https://ot.aem.live |
| Resource loading | Script, CSS, image load times | https://ot.aem.live |
| Errors | Client-side JavaScript errors | https://ot.aem.live |
RUM data is collected via sampleRUM() in aem.js and sent to Adobe's Helix analytics endpoint.
Custom Data Layer
| Component | Events Tracked |
|---|---|
| All blocks with analytics config | data-trackinview (IntersectionObserver), data-trackclick (delegated) |
| Cookie Consent Banner | ccb-shown, ccb-hidden, ccb-close-clicked, ccb-settings-clicked |
| Modal | modal-open:{id}, modal-close:{id}, modal-confirm:{id} |
| ISI | isi:state-change |
| CTA | cta-click |
All events are pushed to window.dataLayer.events with timestamps.
Health Checks
| Check | Method | Frequency |
|---|---|---|
| Live site availability | HTTP GET to aem.live URL | Continuous |
| Preview availability | HTTP GET to aem.page URL | On-demand (pre-publish) |
| AEM Author availability | Adobe status page | Continuous |
| Lighthouse performance | Automated Lighthouse CI | Per PR (via PSI checks) |
Where the logs go
The three sections above are application-level telemetry. Infrastructure logs are routed separately:
| Traffic Type | Log Source | Destination |
|---|---|---|
| Edge delivery traffic | Cloudflare Logs | Splunk (via Cloudflare Logpush) |
| AEM runtime / application | Adobe AEM Cloud Logging | New Relic / AEM Log Forwarding |
| Worker errors | Cloudflare Worker Logs | Cloudflare dashboard / Logpush |
Cloudflare is the primary source for request-level visibility because it sees incoming client traffic directly. AEM/New Relic observability covers the AEM runtime and application layer.
Cloudflare logs only exist for environments that are actually Cloudflare-fronted — Environment Access Control lists which those are.
Common Issues and Resolution
| Issue | Likely Cause | Resolution |
|---|---|---|
| Block not rendering | Missing block JS/CSS in repository | Verify block files exist, check console for 404 errors |
| Stale content on live | Content not published via Sidekick | Publish the page using Sidekick |
| Preview shows old content | Browser cache | Hard refresh or clear Sidekick cache |
| Broken layout on mobile | Missing responsive styles | Check SCSS breakpoints, verify @include media() usage |
| ISI not appearing | Missing page metadata or ISI fragment | Verify enableIsi and isiPath metadata |
| Theme toggle not working | theme-config.json misconfiguration | Verify showToggler: true and onlyLightMode: false |
| Fragment not loading | Incorrect fragment path | Verify fragment exists at the expected path in AEM author |
| Form submission failing | CORS or endpoint issue | Check form action URL and server-side endpoint |
Related Documents
- Onboarding Guide — Onboarding and rollback procedures
- Content Authoring Model — Content authoring workflow
- Security Model — Security incident response
- DevOps Overview — Pipeline and delivery model