Is there a way to add layers through the lambda deployment pipeline?

[Kyle Godfrey]

Lots of the functions we'll be creating will share various node libraries so it would be helpful to specify layers to add to the versions. That way we could more tightly control library versions over time while have a better consistency across functions. Currently I'm not seeing any documentation on if layers are controllable through the lambda pipeline. Would the only option be to combine this pipeline with SAM CLI (also not sure if this is possible)?

[Halyna Berezovska · Accepted solution]

@Kyle Godfrey regarding SAM CLI, you can use the Bitbucket Pipe in your pipeline. Check it out if it cover your needs https://bitbucket.org/atlassian/aws-sam-deploy/src/master/README.md .

Regarding layers, please give us more details about what exact layers you mean? If you mean lambda layers, this depend on your AWS lambda setup https://docs.aws.amazon.com/lambda/latest/dg/configuration-layers.html#configuration-layers-create  .

Also you can manage layers when updating lambda configuration and use aws-lambda-deploy pipe passing FUNCTION_CONFIGURATION variable.

More clear option IMHO is to use aws-sam-deploy pipe with a proprer sam_template.yml template. Check out how you can mamange layers in sam template https://docs.aws.amazon.com/serverless-application-model/latest/developerguide/serverless-sam-cli-layers.html .

Constructing proper sam_template.yml , you will pass it as a variable SAM_TEMPLATE to aws-sam-deploy pipe (check out its README).

Choose the option , you like most.

Looking forward to hearing from you, if that covers your requirements and work or not for you.

Regards, Galyna