'What happens with AWS Aurora Read Replica in case of AZ failure

I can't seem to find anywhere the answer to this question.

We have an AWS Aurora cluster with a R/W instance and a read replica (multi-AZ). What happens with the read replica in case of a failure of it's AZ?

What would the DR plan be?

Will AWS create a new read-replica in a new AZ (because we have configured it as multi-AZ)? Do we need to create a second read-replica in another AZ just to factor in for this kind of failure?



Solution 1:[1]

I believe that I found the answer to this question.

The Aurora Auto scaling with Aurora replicas feature set to minimum 1 replica would fix this problem. This will ensure that a read-only replica will be provisioned in a different AZ in case when an AZ is offline.

Please correct me if this is wrong.

Sources

This article follows the attribution requirements of Stack Overflow and is licensed under CC BY-SA 3.0.

Source: Stack Overflow

Solution Source
Solution 1 Dan Nemes