Case study 06 / 06
Cloud infrastructure · AWS ECS · Resilience
Cross-Region Disaster Recovery & Reliability
Cross-region AWS ECS disaster-recovery readiness, with configuration-managed endpoint switching that repoints domains during DR, plus service health checks and monitoring.
- Role
- Contributor
- Company
- ICICI Lombard
- Stack
- AWS ECS
- Docker
- Health checks
- Monitoring
- Configuration as code
01
Overview
Disaster-recovery and reliability work for services running on AWS ECS across regions.
02
Context
Failover that depends on manual changes under pressure is slow and error-prone.
03
Problem
Make recovery a configuration change rather than a code change, and know whether services are healthy before and after a switch.
04
My contribution
- Worked on cross-region ECS disaster recovery and deployment readiness.
- Contributed to configuration-driven DR endpoint switching that automatically repoints domain-level endpoints during disaster recovery — no extra code written at failover time. Live in production for RYD Raksha and two-wheeler.
- Contributed to service health checks and health monitoring.
- Supported fault-tolerant deployment and production troubleshooting.
05
Architecture
Two regions, one switch
Services run on ECS in a primary region with a recovery region ready. Endpoint configuration lives in code; during DR, domain-level endpoints are repointed to the recovery region by configuration, and health checks confirm service state.
- Containerized services on AWS ECS (Docker).
- Primary and recovery regions.
- Configuration-managed endpoint switching.
- Health checks and monitoring around the switch.
06
Impact
- No RTO, RPO or availability figures have been measured for publication, so none are claimed.
Described at a public level — no confidential company, customer or system details.