Adding Environment Variables to Serverless Functions

2022-09-10 282 words 2 minutes

Environment variables can be set for each individual Serverless function or at the provider level for all functions.

Set Environment Variables for Individual Functions

To add environment variables to a Serverless function, define them using environment under the function name. In the example below, the function function1 will have the environment variable S3_BUCKET_NAME defined:

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
provider:
  name: aws
  runtime: python3.9
  stage: production
  region: us-west-1

functions:
  function1:
    handler: function.lambda_handler
    environment:
      S3_BUCKET_NAME: bucket1 # setting environment variable

Set Environment Variables for All Functions

It’s also possible to add environment variables to all functions in the Serverless template by defining variables under provider:. In the example below, the functions function1 and function2 will have the environment variable S3_BUCKET_NAME defined:

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
provider:
  name: aws
  runtime: python3.9
  stage: production
  region: us-west-1
  environment:
    S3_BUCKET_NAME: my-bucket # setting env variable for all functions

functions:
  function1:
    handler: function1.lambda_handler
  function2:
    handler: function2.lambda_handler

Combining Both Ways of Setting Environment Variables

Finally, it’s possible to add environment variables under provider as well as under each function.

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
provider:
  name: aws
  runtime: python3.9
  stage: production
  region: us-west-1
  environment: # env variables for all functions
    STAGE: production
    REGION: us-west-1

functions:
  function1:
    handler: function1.lambda_handler
    environment:
      S3_BUCKET_NAME: bucket1 # setting env variable for only function1

  function2:
    handler: function2.lambda_handler
    environment:
      TABLE_NAME: table # setting env variable for only function2

In this case, both function1 and function2 will have both the STAGE and REGION environment variables set, but function1 will have S3_BUCKET_NAME set while function2 will have TABLE_NAME set.