Maintenance: add canary step to layer workflows

[am29d]

Summary

During the current layer distribution we run e2e tests before deploying the layer. This gives us confidence that layer deployment will work. However, after we deploy a public layer we do not run canary tests to test this specific layer. Thus, we should add a canary step to our workflow, similar what powertools python have, so we can catch any layer deployment issues during a release.

Why is this needed?

There is a chance that the public layer we have distributed does not work, and we have no mechanism right now to catch this problems.

Which area does this relate to?

No response

Solution

Add a canary stack, see canary_stack from powertools python, that adds the deployed public layer and runs a lambda function using powertools. We should roll out canary deployment after beta and prod deployments in our layer workflow.

Acknowledgment

• This request meets Lambda Powertools Tenets
• Should this be considered in other Lambda Powertools languages? i.e. Python, Java, and .NET

Future readers

Please react with 👍 and your use case to help us understand customer demand.