Why set max_azs=2 on RdsProxyStack vpc ?

[anentropic]

I am building a stack based on "The RDS Proxy" in this repo.

I noticed here: https://github.com/cdk-patterns/serverless/blob/main/the-rds-proxy/python/the_rds_proxy/the_rds_proxy_stack.py#L19C36-L19C43

...the VPC has max_azs=2 set.

I just wanted to understand the reason for that, and if I need to put more thought into that value for my own case.

Is it because typically RDS multi-az will span 2 AZs and the VPC only needs to cover the AZs where the db instances are located?

In that case could/should I set max_azs=1 if I am doing a single-az RDS instance?

Similarly, I think my region has 3 AZs, and RDS multi-az may span up to 3 AZs: https://aws.amazon.com/rds/features/multi-az/#Amazon_RDS_Multi-AZ_with_two_readable_standbys ...so I guess in that case I'd set max_azs=3 on the VPC?

multi_az is a boolean flag on aws_rds.DatabaseInstance so it's not very clear to me how you control how many AZs you want for the db, unless that is inferred from the VPC?