RDS endpoint name format

Question

In AWS, RDS Endpoint names have the following format

my-db-name.aaaaaaaaaaaa.eu-west-1.rds.amazonaws.com

the second part of the identifier is a series of seemingly random characters. How is this generated? It seems to be consistent across all rds instances in the same account and region. Are there any other factors that would cause it to change.

I had a look thorugh the AWS docs but couldn't find any relevant information.

Answer 1

It doesn't seem to be clearly addressed in the documentation, but there are AWS support forum posts written by AWS employees, like the one below.

In summary, it's an opaque identifier, internally generated by an algorithm that isn't publicly disclosed, and it should not change for a given region+account combination. A new instance, reusing an old name, should have the same endpoint as the old instance.

The term "hexchars," here, refers to the component of the endpoint hostname that you're asking about:

1. Is there some way to "predict" or otherwise determine what the hexchars will be?

You can't predict it, but once you create at least one instance in a region you will know what it is for your account in that region (and it will be the same for all RDS instances belonging to the same account in that region).

2. Is there a way to set this?

No. The "hexchars" component is generated internally and is unique to your AWS account in each region (cannot be changed). The "hexchars" will be different for the same account in a different region.

3. Is the hexchars consistent if the instance ID is consistent?

Yes. You can delete an instance and if you create another with the same name, it will have the same endpoint address as the original instance (for the same AWS account in the same region). Also, if you rename an instance from "myinst1" to "myinst2", the first component is the only that changes and the "hexchars.region.rds.amazonaws.com" remains the same.

— https://forums.aws.amazon.com/thread.jspa?messageID=486170&#486199