d3.micro Postgres RDS resetting instance every day

Dear Customer,

I thank you for reaching out to us via AWS re:Post.

I think, to deep dive on the instance issues a support case to the Premium Support Team would help as they can critically analyze the issue and provide you the result.

[] Monitoring Amazon RDS API calls in AWS CloudTrail - https://docs.aws.amazon.com/AmazonRDS/latest/UserGuide/logging-using-cloudtrail.html

Apart from that, I would suggest, checking CloudTrail for any app/unwanted-user triggering the restarts(or maybe some automation) - I sincerely hope I was able to shed some light on the matter for you.

This is a very perplexing situation, as RDS instances are designed for persistence and do not reset or delete databases on their own. The behavior you're describing almost certainly points to an external automated process that is replacing or reverting your RDS instance on a daily schedule. Since you are not seeing any details in the standard database logs, it is highly likely that the operation is happening at the AWS infrastructure level, not through a DROP DATABASE command inside Postgres.

The most probable cause is an automated script or a CI/CD pipeline performing a Point-in-Time Restore or restoring from a snapshot. This action would replace your current instance with an older version, effectively "deleting" all the changes made since that snapshot was taken, including your new database and its data. Another common culprit is an Infrastructure as Code (IaC) tool like AWS CloudFormation or Terraform. If there's a daily scheduled job that applies a configuration from one of these tools, and that configuration defines the RDS instance without your new database, the tool could be reverting the instance to its original state every day.

To find the root cause, your most critical tool will be AWS CloudTrail. You must check the CloudTrail event history for your account, specifically filtering for API calls related to your RDS instance. Look for events such as:

• RestoreDBInstanceToPointInTime
• RestoreDBInstanceFromDBSnapshot

DeleteDBInstance followed by a CreateDBInstance The details in CloudTrail will show you exactly which user or automated role is making these calls and from where, allowing you to trace the source of this daily reset. Investigating CloudTrail first is the fastest way to solve this mystery.