We recently started using Aurora MySQL, including a read replica. We implemented an auto scaling policy. The policy worked successfully, the read replica came on board, offloaded the connections, then as things calmed down, the connections drained and it was removed.

All told this process took 30 minutes.

Site 24x7 never picked up the auto scaling RDS instance. I contacted support and the RDS polling time is hourly. Unless the auto scaling instance lasts for an hour, at the minimum, it will not be displayed.

I would like to hereby request a new feature where polling for RDS instances is considerably lower to pick up auto scaling instance, perhaps 5 minutes or less. I would also like to see some sort of persistence in the metrics for those instances. For example, I'd like to be able to click on the read replica, see that it spiked CPU, and see that an auto scaling instance relieved the pressure, then was removed.

Thanks