Deploy a Function and Verify It Runs
A 3-step recipe with AWS Lambda. We can help you get access.
Recipe · 3 steps
Deploy a Function and Verify It Runs
with AWS Lambda · IT Infrastructure Management
Create a Lambda function, wait for it to become Active, and test-Invoke it.
- Create the Lambda function from the staged S3 deployment package. Publish is left false so the unpublished $LATEST version is what gets tested; a version is only cut once the code is known good.
- Poll the function configuration until Lambda finishes provisioning. A newly created function reports State "Pending" first; invoking before it reaches "Active" is rejected, so this step retries until the state settles.
- Invoke the function synchronously with the test payload. The Tail log type returns the last 4 KB of the execution log base64-encoded, and the X-Amz-Function-Error response header is present when the handler itself threw, which a 200 status
What you'll need to have ready
- functionName: The name of the function to create (e.g. order-processor).
- runtime: The runtime identifier for the function (e.g. nodejs20.x, python3.12).
- roleArn: The ARN of the IAM execution role Lambda assumes when running the function.
- handler: The entry point in the code (e.g. index.handler).
- s3Bucket: The S3 bucket holding the deployment package ZIP.
- s3Key: The S3 key of the deployment package ZIP.
- description: An optional human-readable description for the function.
- memorySize: Memory in MB allocated to the function.
- timeout: Maximum execution time in seconds before Lambda stops the invocation.
- environmentVariables: Map of environment variable name/value pairs to expose to the handler.
- testPayload: The JSON event passed to the function during the verification Invoke.
For your developer
Arazzo workflow aws-lambda:deploy-function, 3 operations: CreateFunction → GetFunctionConfiguration → Invoke. Specs on APIs.io.