Abstract

High availability and disaster recovery are critical aspects of cloud storage. Microsoft Azure provides geo-redundant storage to safeguard against regional failures. In instances where the primary storage region becomes inaccessible, unplanned failover enables organizations to designate a secondary region as the new primary, ensuring continued data access. This paper examines the mechanisms, implications, and challenges of unplanned failover and failback in Azure Storage. Due to asynchronous replication, unplanned failover may result in data loss, rendering the Last Sync Time metric essential for assessing replication status prior to initiating failover. Furthermore, as failback is not automated, organizations must manually restore geo-redundancy and reconfigure replication settings, incurring additional costs. To optimize failover readiness, enterprises must implement proactive monitoring, automated alerts, and application resilience strategies. By integrating failover into a comprehensive disaster recovery plan, organizations can minimize downtime, mitigate data loss, and ensure business continuity in the event of regional disruptions.