Select Language:
If you’ve been waiting a long time for your RDS snapshot to export to S3 and it’s still stuck in the “Running” state, you’re not alone. This can happen even with smaller databases, like an 11 GB MySQL Aurora instance. Usually, the export process goes through a few steps—restoring the database, scaling it, then extracting data to your S3 bucket. Sometimes, the process seems to stall at 99%, even though you see files being created and then deleted in your S3 bucket. This indicates the process is still working, but it might be running into issues.
A few things can impact the time it takes to complete the export. For example, databases with well-organized, numeric primary keys tend to export faster than those with string-based keys or lacking good partitioning. Also, since the process requires restoring the entire database first before extracting data, larger or more complex databases may take longer.
Here are some steps you can follow to troubleshoot this problem:
1. Make sure your IAM role has the correct permissions. It should allow the “export.rds.amazonaws.com” service principal to assume it properly.
2. Check that your S3 bucket is in the same region as your RDS database. Exporting across regions can cause delays and errors.
3. Look into your S3 bucket policy to ensure nothing is blocking access or interfering with the export.
4. Use the AWS Management Console, AWS CLI, or RDS API to monitor the export task. This can give you more detailed information about where the process might be stuck.
If these steps do not resolve the issue, consider the following options:
– Cancel the current export and start a new one.
– Use AWS Database Migration Service (DMS) as an alternative to export your RDS data to S3.
– Contact AWS Support for help specific to your account and setup.
Also, the “Delete marker” you see for some files is usually part of the normal process. It happens when the service checks write permissions to your S3 bucket before moving forward with the export.
In short, if you see your export stuck at 99%, it’s likely due to a permission or regional setting issue or a related process taking longer than usual. Going through these troubleshooting steps can help you pinpoint the problem and get your data exported successfully.





