- Newest
- Most votes
- Most comments
1. Upgrade Timing & Phased Rollout
When an engine reaches End of Standard Support and Extended Support is disabled (open-source-rds-extended-support-disabled), AWS does not force upgrades universally on the exact EOL date. Forced lifecycle upgrades are executed in a phased global rollout during configured maintenance windows. This process routinely takes several months, which explains the gap between February and August.
2. Target Version Selection (Directly to 18.3)
To prevent consecutive disruption and avoid upgrading an EOL instance to a version that is also nearing its own EOL (e.g., PG 14 or 15), AWS forces upgrades directly to a newer, long-term default major version. Both PostgreSQL (pg_upgrade) and AWS RDS natively support multi-major version direct upgrades, provided the required minor version baseline (e.g., 13.23) is met first.
3. Absence of CloudTrail Events
CloudTrail records API requests made by IAM principals (customers). Because this forced upgrade was a system-initiated action executed internally by the RDS Control Plane, no ModifyDBInstance API call was generated. System-enforced lifecycle events are pre-announced via the AWS Health Dashboard and logged solely in the RDS Event History.
Official AWS Documentation References:
- RDS Extended Support & Automatic Upgrades: https://docs.aws.amazon.com/AmazonRDS/latest/UserGuide/extended-support.html
- Behavior of Deprecated Engine Versions (Forced Upgrades): https://docs.aws.amazon.com/AmazonRDS/latest/UserGuide/USER_UpgradeDBInstance.Upgrading.html#USER_UpgradeDBInstance.Deprecated
- Upgrading PostgreSQL Major Versions: https://docs.aws.amazon.com/AmazonRDS/latest/UserGuide/USER_UpgradeDBInstance.PostgreSQL.html
Relevant content
asked 3 years ago

If my answer was helpful, I would appreciate it if you could mark it as the accepted answer.