RDS Instances

Uptime Scheduler starts and stops RDS DB instances using the native AWS start/stop APIs. All data is preserved between cycles.

Supported states

State	Meaning
available	Instance is running and accepting connections
stopped	Instance is stopped (storage still billed, compute is not)
starting	Starting up
stopping	Shutting down

How scheduling works

• At schedule start time — rds:StartDBInstance is called. The instance moves through starting → available.
• At schedule end time — rds:StopDBInstance is called. The instance moves through stopping → stopped.

RDS start/stop operations can take several minutes. The scheduler monitors the instance status and waits for the operation to complete.

Tag example

uptime:schedule = 0800-1800 weekdays
uptime:env      = dev

Supported engines

All RDS engines that support the stop/start API are supported:

• MySQL / MariaDB
• PostgreSQL
• Oracle
• SQL Server

Amazon Aurora is not currently supported — Aurora clusters use a different API and cluster-level start/stop is not yet implemented.

Multi-AZ

Multi-AZ instances are fully supported. The stop/start APIs operate on the primary instance; the standby is handled automatically by RDS.

AWS-imposed limits

AWS automatically starts a stopped RDS instance after 7 days to apply pending maintenance. If your resource has a schedule that covers the 7-day window, this is not an issue. If you stop an RDS instance outside a schedule window manually, be aware that AWS may restart it.

Parameter groups and option groups

These are not affected by stop/start cycles.