Skip to main content
Version: v1.4

Disaster Recovery and Failover

Nova can support disaster recovery workflows by orchestrating recovery actions across clusters in response to failure events.

When to Use This

Use this pattern when:

  • Workloads need to recover from cluster, zone, or region failures
  • Recovery actions should be automated
  • Applications need to move, restart, or be reconfigured during failure events
  • Recovery logic should be defined declaratively

How Nova Helps

Nova provides a Kubernetes-native recovery framework that enables automated response to failure events.

Recovery behavior is driven by:

  • Incoming alerts
  • Recovery plans that define the required response
  • Automated execution of recovery steps

This allows recovery workflows to be expressed declaratively and executed by the Nova control plane.

How It Works (High-Level)

At a high level, Nova:

  1. Receives an alert from an external system
  2. Matches the alert to a defined recovery plan
  3. Executes the recovery steps

For details on how recovery is implemented, including the underlying resources and workflow, see:

Recovery Framework

Considerations

Nova orchestrates recovery actions, but application-level recovery still depends on:

  • Data replication
  • Storage availability
  • DNS and traffic routing
  • Application readiness
  • External dependencies