Foundationdb has a good solution to this. Uses three regions, sync writes to regions 1 and 2 which should be near each other, sync writes to region 3. If region 1 dies then region 3 pulls last commits from region 2 before serving traffic. Small cluster manager runs in all three regions
But the scaling problem is still there, when they switched over I assume they would start both horizontal and vertical scale of k8 pods and maybe even bump up DB instances as well which could result in a small congestion at the start of the switch over