Automating for Failure: Disaster Recovery Testing with Terraform & Gremlin
summary
Claims an automated DR pipeline that actually spins a live replica after failure using Terraform + Gremlin for tests and AWS for recovery. Production EC2 state changes hit EventBridge → CodeBuild → Terraform, CodeBuild runs terraform apply to recreate VPC, EC2 and web server in a separate account and region, and the demo shows the site returning. Notes Terraform does not back up data, cross-account event bus needs manual wiring today, and fire drills are required.
AI-written summary · about
chapters (9)
- 0:00Introduction to speakers
- 0:46Why prepare for failure
- 3:35Backup vs disaster recovery
- 7:32RTO and RPO explained
- 9:16Terraform for disaster recovery
- 11:42Demo setup overview
- 17:45Configuring EventBridge rules
- 23:23Triggering recovery with CodeBuild
- 25:32Key takeaways and conclusion
related talks
28mAutomating Chaos Engineering GameDays with Terraform
52mTest Driven Delivery of Infrastructure-As-Code
47mBuilding Resilience
35mMulti Region Terraform Deployments
31mMaking AI work for you: Terraform engineer blueprint
57mLook Ma, no OPS! Automating microservices' infrastructure
35mMisadventures with Terraform
36mTerraform & CI/CD