Answer: B
Option A is incorrect because Multi-AZ with Automatic Failover must first be disabled on an ElastiCache (Cluster Mode Disabled) cluster to promote a read replica node to a primary node manually.
Option B is CORRECT because Multi-AZ with Automatic Failover must first be disabled on an ElastiCache (Cluster Mode Disabled) cluster to promote a read replica node to a primary node manually.
Option C is incorrect because creating a manual backup is not required. Instead, Multi-AZ with Automatic Failover must first be disabled on an ElastiCache (Cluster Mode Disabled) cluster to promote a read replica node to a primary node manually.
Option D is incorrect because it is not possible to stop an ElastiCache cluster.
Reference:
https://docs.aws.amazon.com/AmazonElastiCache/latest/red-ug/AutoFailover.html