Automatically provisioning an instance to connect to a database cluster endpoint - amazon-web-services

Question

I am looking for a way to provision an instance with a configuration file that contains the endpoints to connect to a database cluster in an automatic way, using terraform. I am using a aws_rds_cluster resource, from which I can get the endpoint using the expression aws_rds_cluster.my-cluster.endpoint. Then, I would like to provision machines instantiated with an aws_instance resource so that the value of that expression is stored in the file /DBConfig.sh.
The content of the DBConfig.sh file would look like this :
#!/bin/bash
ENDPOINT=<$aws_rds_cluster.my-cluster.endpoint$>
READER_ENDPOINT=<$aws_rds_cluster.my-cluster.reader_endpoint$>
Truth be told, once I successfully reach that point, I'd like to be able to do the same thing for machines created by a aws_launch_configuration resource.
Is this something that can be done with terraform? If not, what other tools can I use to achieve this kind of automation? Thanks for your help!

Answer 1

There are few ways which can achieve that. I think all of the would involve user_data.
For example, you could have aws_instance with the user_data as follows:
resource "aws_instance" "web" {
# other atrributes
user_data = <<-EOL
#!/bin/bash
cat >./DBConfig.sh <<-EOL2
#!/bin/bash
ENDPOINT=${aws_rds_cluster.my-cluster.endpoint}
READER_ENDPOINT=${aws_rds_cluster.my-cluster.reader_endpoint}
EOL2
chmod +x ./DBConfig.sh
EOL
}
The above will launch an istance which will have DBConfig.sh with resolved values of the endpoints in its root (/) directory.