Incident response & recovery

Executive Summary

Eli maintains a streamlined incident response approach leveraging our cloud providers' professional-grade monitoring and support capabilities. Our response framework focuses on rapid provider escalation and clear customer communication to minimize impact and ensure quick resolution.

Incident Detection & Response

Automated DetectionOur cloud providers automatically monitor and alert on:

  • Database issues - MongoDB Atlas performance, connectivity, or backup problems

  • Application errors - Vercel deployment failures or runtime issues

  • Authentication disruptions - Firebase Auth service problems

  • Email delivery failures - Resend delivery issues

Manual Detection

  • Customer reports via email or support channels

  • Team monitoring during business hours (9 AM - 6 PM Paris time)

  • Provider status page reviews for ongoing service issues

Response Procedures

Service Provider Issues

  1. Identify affected provider through alerts or customer reports

  2. Check provider status pages for known incidents and estimated resolution

  3. Escalate to provider support if issue is not already acknowledged

  4. Monitor provider updates and communicate timeline to affected customers

  5. Verify service restoration once provider reports resolution

Application Code Issues

  1. Assess impact and identify affected functionality

  2. Implement hotfix through standard deployment process

  3. Deploy via Vercel with automatic rollback capability if needed

  4. Monitor application post-deployment for resolution confirmation

  5. Document issue for future prevention

Provider Support Access

Professional Support Coverage

  • MongoDB Atlas - Professional support with 4-8 hour response times for critical issues

  • Vercel - Direct support access through web and email channels

  • Firebase/Google Cloud - Professional support with 4-hour response for high-impact issues

  • Resend - Direct chat support for email delivery issues

Customer Communication

Communication Approach

  • Proactive notification for service-wide issues affecting multiple customers

  • Direct customer contact for organization-specific problems

  • Transparent updates including provider status and estimated resolution times

  • Post-resolution summary for significant incidents

Communication Channels

  • Email to customer administrative contacts

  • Provider status page references for infrastructure-related issues

Business Continuity

Service Recovery Priorities

  1. Database connectivity - MongoDB Atlas failover and connectivity restoration

  2. Application availability - Vercel service restoration and deployment capability

  3. User authentication - Firebase Auth service functionality

  4. Email delivery - Resend service restoration for notifications

Continuous Improvement

  • Post-incident review of response effectiveness and customer communication

  • Provider relationship management ensuring appropriate support access

  • Process documentation updates based on actual incident experience


For incident reporting or service issues, contact: [email protected]

Last Updated: August 2025

Mis à jour