From the last article, I wanted to get Version Control for my work so far and to get the contents of the S3 bucket served through CloudFront. Both of these items have been accomplished. Git Repository is available on GitHub. I have made it a public repository. I am fairly confident that I can keep secrets out of it. CloudFront is now serving the S3 bucket with a nice url: ToMakeItGo!

Things of note for this portion of the adventure. I resorted to using the management portal for handling the Certificate and DNS configurations. I don’t currently have my domain registered with Amazon. I didn’t get the key generation scripted out. This has lead to one weird line where I have to reference the key in stack file by its ARN (Amazon Resource Name, Identifier to identify an amazon resource).

        ViewerCertificate:
          SslSupportMethod: sni-only
          AcmCertificateArn: !Join
            - ''
            - - 'arn:aws:acm:us-east-1:'
              - !Ref AWS::AccountId
              - ':certificate/36c80ae0-8618-4221-a636-039990e65e0a'

I had to use that !Join list of lists syntax to get access to the key defined outside of the stack file. I was able to use a variable, so I don’t have my account id in version controlled file. I don’t think I am leaking too much information here. us-east-1 is the only supported region for an ACM Certificate that is allowed in this usage. The GUID without the account is unlikely to be useful.

These macros are a bit odd for me to see, but I think I am still on board with this route for configuration. The few I have used have been text replaces. I don’t think there is any jumps or iterations to make things hard to follow.

Short term I would like to get the build of Jekyll blog more automated and fill in my README.md file.

I have yet to figure out what the next step in this journey of AWS should be. Likely it will involve some sort of API Gateway. A simple call and response call like an echo might be a good spot to start.