A bank's transactional database uses active geo-replication to a secondary in another region. During a regional failure, the team performs a forced failover and notices that a few recent commits are missing from the new primary. Auditors ask why this happened and how the bank can reduce the risk in future. Which statement is the most accurate explanation?
Show answer and explanation
Answer: A
Active geo-replication sends committed transactions to the secondary asynchronously. A commit can be acknowledged on the primary before its log records reach the secondary, so a sudden regional loss can leave a small gap. Azure SQL does not offer a synchronous cross-region commit for this topology. The bank can reduce exposure by performing planned failovers while the primary is healthy, because those synchronize first, and by monitoring replication lag so that any potential gap is known in advance. The other explanations do not describe how replication works.
Why the other options are wrong
- B. Active geo-replication replicates the full database, including data changes, not just the schema. Missing recent rows points to replication lag at the moment of failure rather than to a schema-only feature.
- C. The geo-secondary is readable and receives changes from the primary. It does not reject writes that were already replicated, so losing commits is about replication timing rather than a read-only mode mismatch.
- D. Zone redundancy keeps replicas within one region across availability zones. It does not write commits into a different region's transaction log, so it cannot explain a cross-region gap.