Posted On: Apr 4, 2017
We are happy to announce that ElastiCache for Redis now allows customer-triggered automatic failover. In clusters that have one or more read-replicas with Multi-AZ enabled, you can trigger a failover by manually failing the primary. This will initiate automatic failover, which will promote a read replica to primary, and replace the failed primary as a new read replica in place of the one promoted. Triggering auto failover can help with testing how your application responds to a failure of an ElastiCache primary.