I'm having trouble with my RDS reserved instances. I have two database instances: a db.r7g.2xlarge, which is my master Aurora MySQL, and a db.r7g.large, which is a read-only instance for processing. I purchased reserved instances for both sizes, but I'm seeing some unexpected billing. It seems like the system is fully utilizing the large reservation on the 2xlarge instance and charging me an additional $320 per month, while only partially using the 2xlarge reservation on the larger instance. I've reached out for support, but I've been told that this behavior is normal, which only confuses me more. It feels like a bug because the reserved instances aren't matching up properly with my actual usage. Anyone else experienced this?
3 Answers
Have you checked if your instances are Multi-AZ? Sometimes that can affect how reserved instances are applied. Just a thought!
Make sure your reserved instances are set to Regional Scoped with Size Flexibility. This helps apply them across multiple instances in the region. If you're fixed on specific sizes, then you might need an exact match for both your r7g.2xlarge and r7g.large. AWS tends to apply these reservations dynamically, so keeping everything aligned can be tricky!
I think I did that already. Everything matches, but it’s still not utilizing the reservations the right way.
It might be worth researching further into the billing side as well. There could be some nuances in how AWS is processing the reservations that aren't immediately clear. Make sure to check if other users are having similar experiences!
Actually, Aurora doesn’t really use Multi-AZ the same way as others—it’s more about how each instance is set up. Just make sure your RIs are scoped by region.