EC2 instance - Elastic IP disassociated overnight

The issue you're experiencing with the Elastic IP address being disassociated from your EC2 instance overnight is unusual and not a standard behavior for Elastic Beanstalk or EC2 instances. Since there are no disassociation events in CloudTrail and no manual changes were made, it's challenging to pinpoint the exact cause without more information.

However, there are a few potential reasons and steps you can take to investigate and potentially resolve this issue:

1. Check your Elastic Beanstalk environment configuration: Ensure that there are no scheduled actions or configuration changes that might be triggering instance replacements or updates overnight.

2. Review your EC2 Auto Scaling settings: If your Elastic Beanstalk environment is using Auto Scaling, check if there are any scaling activities occurring overnight that might be terminating and replacing instances.

3. Examine EC2 instance status checks: Look for any failed status checks or instance recoveries that might have occurred, which could potentially lead to instance replacement.

4. Monitor your Elastic IP address associations: Set up CloudWatch alarms to notify you when an Elastic IP address is disassociated from an instance.

5. Check for any AWS service issues: Review the AWS Service Health Dashboard for any reported issues with EC2 or Elastic Beanstalk in your region during the times when the disassociation occurs.

6. Consider using an Elastic Beanstalk operations role: This feature (currently in beta) allows Elastic Beanstalk to manage permissions for downstream services, which might help maintain more consistent resource associations.

If the issue persists, contact AWS Support: They can help investigate the root cause by examining logs and events that might not be visible to you in the console.

As a temporary workaround, you can create a script that periodically checks the Elastic IP association and re-associates it if necessary. However, this should be considered a short-term solution while you work on identifying and addressing the root cause.

Remember to document each occurrence, including exact times and any relevant instance or environment information, as this will be valuable for troubleshooting and if you need to escalate to AWS Support.
Sources
Migrate to a new instance type by launching a new EC2 instance - Amazon Elastic Compute Cloud
DisassociateEnvironmentOperationsRole - AWS Elastic Beanstalk