Question

A retail company runs a stateless containerized web API on Azure Kubernetes Service (AKS) in the East US region. The business requires the API to survive a full regional outage with an RTO of under 5 minutes and an RPO of zero (the workload holds no persistent state; all data lives in an externally managed Azure Cosmos DB account configured for multi-region writes). The solution must automatically route users away from a failed region without manual intervention and must minimize ongoing cost where possible. Which approach best meets these requirements?

Accepted Answer

Deploy identical AKS clusters in East US and West US, place both behind Azure Front Door with health probes and priority/latency routing, and let Front Door fail traffic over automatically.

Answer

Deploy a single AKS cluster spread across three availability zones in East US and rely on zone redundancy for disaster recovery.

Answer

Deploy the AKS cluster only in East US and configure Azure Site Recovery to replicate the cluster's node VMs to West US, failing over the VMs during a regional outage.

Answer

Deploy AKS clusters in East US and West US and use Azure DNS with a manually updated CNAME record to point to the healthy region when an outage occurs.