Examkin

Question

Difficulty: HardDatabase High Availability

A company's production database runs on an Amazon RDS for MySQL DB instance configured in a Multi-AZ deployment. The primary DB instance is located in the us-east-1a Availability Zone, and the standby DB instance is in us-east-1b. To support disaster recovery, a cross-Region Read Replica has been deployed in the us-west-2a Availability Zone. An Availability Zone outage occurs in us-east-1a, causing the primary DB instance to become unreachable.

Which of the following describes the automatic failover behavior of the database tier and the required action from the SysOps Administrator?

  1. A
    Amazon RDS promotes the standby instance in us-east-1b to primary, but the replication to the cross-Region Read Replica in us-west-2a is broken. The administrator must manually promote the Read Replica to a standalone instance, re-establish replication from the new primary, and demote it back to a replica.
  2. B
    Amazon RDS promotes the cross-Region Read Replica in us-west-2a to be the primary instance because it is in a separate Region. The standby instance in us-east-1b is then converted to a Read Replica, and the administrator must manually update the application connection strings to point to us-west-2a.
  3. Amazon RDS automatically updates the DNS CNAME record of the DB instance endpoint to resolve to the standby instance in us-east-1b. The cross-Region Read Replica in us-west-2a automatically updates its replication source to the newly promoted primary instance and resumes replication. No manual intervention is required.Answer
  4. D
    Amazon RDS initiates failover, but the administrator must configure an Amazon Route 53 Active-Passive failover routing policy to manually redirect replication traffic from the cross-Region Read Replica in us-west-2a to the standby instance in us-east-1b.

Answer

Amazon RDS automatically updates the DNS CNAME record of the DB instance endpoint to resolve to the standby instance in us-east-1b. The cross-Region Read Replica in us-west-2a automatically updates its replication source to the newly promoted primary instance and resumes replication. No manual intervention is required.
The correct option is correct because when an Availability Zone outage occurs in a Multi-AZ deployment, Amazon RDS automatically detects the failure, promotes the standby instance (in this case, in us-east-1b) to primary, and updates the DNS CNAME record of the DB instance endpoint to point to the new primary. Furthermore, Amazon RDS automatically updates the replication source of any associated Read Replicas (including cross-Region Read Replicas) to the new primary, so replication resumes without manual intervention.

Step-by-Step Solution

1
Analyze the database deployment architecture and identify the type of failover event.
Recognize that the deployment is a standard Multi-AZ DB instance with a cross-Region Read Replica, and the failure of the primary instance in us-east-1a will trigger automatic Multi-AZ failover within the same Region.
Determining the scope of automatic AWS operations helps clarify which parts of the recovery process are managed by Amazon RDS versus what requires manual administrator action.
2
Determine the behavior of the database endpoints during failover.
Identify that Amazon RDS updates the DNS CNAME record for the DB instance endpoint to point to the newly promoted primary instance in us-east-1b (the standby).
This ensures the application, when using the standard database endpoint, automatically reconnects to the new writer without configuration changes.
3
Determine the behavior of the cross-Region Read Replica.
Verify that Amazon RDS automatically updates the replication source of the cross-Region Read Replica to target the new primary instance in us-east-1b once it is promoted.
This confirms that replication resumes automatically without requiring the administrator to manually rebuild or promote the replica.

Key Concept

Amazon RDS Multi-AZ automatic failover behavior and automatic replication source updates for Read Replicas during failover.
Rate this question

Topics

Database High Availability

More questions from this topic