Which action can the Database Specialist take to test the resiliency of the Aurora DB cluster?
Stop the DB cluster and analyze how the website responds
Use Aurora fault injection to crash the master DB instance
Remove the DB cluster endpoint to simulate a master DB instance failure
Use Aurora Backtrack to crash the DB cluster
Explanations:
Stopping the DB cluster would cause downtime and is not an ideal method to test resiliency. It does not simulate a failure scenario or provide insight into how Aurora handles failover and recovery.
Aurora fault injection allows testing the resiliency of the DB cluster by deliberately causing faults such as instance crashes. This helps verify how well Aurora handles failures and automatic recovery.
Removing the DB cluster endpoint does not simulate a failure of the master DB instance specifically. Aurora uses a DNS endpoint, and the failure of a master instance would result in failover, which isn’t replicated accurately by just removing the endpoint.
Aurora Backtrack is used to revert to a previous point in time, not to simulate a crash or failure. It is not a suitable tool for testing resiliency or the handling of instance failures.
To the best of my knowledge, the answer is:
Use Aurora fault injection to crash the master DB instance
From my point of view, the answer is:
Use Aurora fault injection to crash the master DB instance