I have a lambda function which is pretty general. It queries an Athena Database table, and writes the output to an S3 Bucket. The function is currently set up to accept environment variables for the Database name, table name, output filename and output bucket.

Currently, I call this function with 4 different sets of environment variables, so I have 4 different lambdas, whose code is the same, but whose environment variables are different. Do I need to have 4 different lambdas, or is there a way of having one lambda with 4 environment variable 'sets'?

Thanks!

CodePudding user response：

Here's one option: To handle 4 sets of configuration with a single lambda, send a variable (e.g. type: Foo) part of the lambda invocation¹. As @Marcin suggests, the lambda uses the type value to fetch the config variables from the SSM Parameter Store. Parameters support hierarchies, so you can store your config using nested variable names like: /Foo/db, /Foo/table, /Bar/table etc.

(1) For example, send type in the event detail if event-triggered, or in the SDK Invoke command payload.