How to Really Use AWS Amplify
A quick guide to avoiding *some* of the potholes.
My team and I have had the mixed adventure of using AWS Amplify and its related ecosystem for the past 10 months of our current project. There are many wonderful features that commend this toolset to a variety of developers, but here I am going to discuss its current traps and how to navigate around them.
The Command-Line Interface (Amplify CLI) is not your friend, not yet.
Unfortunately, this tool has some issues that become unwieldy after consistent use. One piece of the toolchain, the amplify push command (documented here), is responsible for parsing your local configuration into a valid AWS CloudFormation template, which it then uses to deploy or update your entire application’s backend in AWS.
Needless to say, this toolchain abstracts a great bunch of complexity out of our hands. However, when issues appear seemingly out of thin air and the toolchain fails to update your stack, you just might be left high and dry with no idea of what to do next. There are many objects and services this command manipulates, and if one happens to throw an error, you might just end up scrapping your entire stack and starting over (we did that twice before getting to production).
