High availability and disaster recovery

The platform is designed for high availability with DR and multi-region Helm profiles. Key points:

• Designed-for is the honest verb. You architect for HA and DR; you do not get to claim an achieved uptime number the deployment has not yet earned. Your design states recovery objectives (RPO and RTO) as targets, with the mechanisms that back them.
• The audit trail and tenant isolation are the parts that must survive a failover intact. Design backup and restore so the tamper-evident chain stays verifiable and the isolation boundary is never relaxed to make recovery easier.
• Use the multi-region Helm profiles as the starting shape, then adapt to the customer's actual failure domains and latency budget.