An Aurora cluster's writer instance fails. What happens to availability and what is the expected failover time?
A.The cluster is down until you manually promote a reader (~30 min)
Failover is automatic; manual promotion is only for DR drills or custom control.
B.Aurora automatically promotes a reader to writer — typically within ~35 seconds
Correct — auto-promotion with the cluster endpoint redirecting writes.
C.A new writer is built from S3 backups (~15 min)
Backup restore would take far longer and isn't the failover path.
D.Failover to a standby in the same AZ only
Aurora replicas span AZs — failover isn't AZ-limited.
Why this answer
Aurora automatically promotes the highest-priority reader to writer and the cluster endpoint follows — typically ~35 seconds. Readers share the cluster storage volume, so promotion doesn't restore from backup.
Correct answer: B · Last reviewed 2026-11-01
Official AWS references
Practice this interactively
Answer without seeing the solution first in the High Availability practice session.
Practice High Availability