Failover Without Fear: A Practical Guide to Active-Active...
summary
Mollie migrates a single-region platform to multi-region on Google Cloud, rewriting application code to be region-aware, moving from the classic to an advanced GCP global load balancer, and adding replica databases (no managed DBs). A staged rollout routes 95%/5% traffic split while pre-warming replicas; first switch on September 26 fails due to an old radius client race condition, teams fix code and tests, then October 9 completes a full switchover following incident-runbook steps and a postmortem.
AI-written summary · about
chapters (7)
- 0:00Introduction
- 2:53Disaster recovery before migration
- 5:45Multi-region architecture
- 13:26Preparation and testing
- 17:21Execution challenges
- 18:29Successful switch-over
- 22:27Key takeaways and lessons learned
related talks
14mThe Multi‑Workload Challenge: Predictable Failover
43mMulti-Cloud Magic: Leveraging Multi-Cloud Clusters in Real-World Scenarios
48mMulti-Cloud Magic: Leveraging Multi-Cloud Clusters in Real-World Scenarios
42mLeveraging Multi-Cloud Clusters in Real-World Scenarios
27mMigrating a Monolith to the Cloud
19mPreparing for disaster: going Multi-Region
40mMulti Cloud Magic: Leveraging Multi Cloud Clusters in Real World Scenarios
36mHow we finally migrated an eCommerce-Platform to GCP