Lambda & S3 Event Notifications - Hands On

This hands-on lab reinforces how AWS utilizes Resource-Based Policies to authorize asynchronous push-model event streams like Amazon S3 notifications without requiring you to embed explicit execution credentials inside the storage layer itself.

🛠️ Step-by-Step S3 Reactive Ingestion Hands On​

1. Provisioning the Code & Storage Buckets​

• Step 1: Bootstrap the Core Lambda Target

  • Spin up a new function named Lambda-S3 utilizing the Node.js runtime environment.

• Step 2: Create a Regionally Co-located Bucket

  • Open the Amazon S3 Console ──► click Create bucket.
  • Name the bucket (e.g., demo-s3-event-rendy). CRITICAL RULE: Ensure the S3 bucket is provisioned in the exact same AWS region as your Lambda function to prevent cross-region latency blocks and complex deployment mappings.

2. Configuring the Push Event Notification System​

• Step 3: Setup the S3 Event Blueprint

  • Navigate to your newly created S3 bucket ──► click the Properties tab ──► scroll down to the Event notifications module.
  • Click Create event notification ──► Name it invoke-Lambda.
  • Leave Prefix and Suffix blank to capture all file streams, or scope it down to target explicit folders.
  • Under Event types, check the box for All object create events (s3:ObjectCreated:*).

• Step 4: Bind the Compute Target Route - Scroll down to the Destination panel, select Lambda function, choose your Lambda-S3 function from the dropdown, and hit Save changes.
  

3. Verification & Live Trace Analysis​

Update your Lambda function code template to print the incoming data dictionary so you can inspect the serialization variables:

export const handler = async (event) => {
  console.log(JSON.stringify(event, null, 2));
  const response = {
    statusCode: 200,
    body: JSON.stringify("Hello from Lambda!"),
  };
  return response;
};

Hit Deploy. Head back to your S3 bucket, upload a sample photo file (e.g., coffee.jpg), and jump over to the Lambda function's Monitor -> View CloudWatch Logs panel.

📥 Deconstructing the Ingested S3 Event Schema Payload​

When you drill into the newly populated CloudWatch log stream, you will find an explicit Records array containing the structural telemetry details emitted by the storage plane:

{
  "Records": [
    {
      "eventVersion": "2.4",
      "eventSource": "aws:s3",
      "awsRegion": "ap-southeast-2",
      "eventTime": "2026-06-25T01:32:15.252Z",
      "eventName": "ObjectCreated:Put",
      "userIdentity": {
        "principalId": "AWS:AIDA24DNY4NTLSCBFI4WZ"
      },
      "requestParameters": {
        "sourceIPAddress": "my-ip-address"
      },
      "responseElements": {
        "x-amz-request-id": "AXES3THZ66BAG331",
        "x-amz-id-2": "VhnsiPGFaT0nTbCQLkit2v8rXNXJOVrrESMP3o2Q4zU8YkK7v/bYP1F6hldKqdFtCfPK8OLZjLVMYOLTsAw8rU9uTgDx+3jZ"
      },
      "s3": {
        "s3SchemaVersion": "1.0",
        "configurationId": "invoke-Lambda",
        "bucket": {
          "name": "demo-s3-event-rendy",
          "ownerIdentity": {
            "principalId": "A18QSH59Y8EC17"
          },
          "arn": "arn:aws:s3:::demo-s3-event-rendy"
        },
        "object": {
          "key": "coffee.jpg",
          "size": 110985,
          "eTag": "b3b29d095d73d905171d1f5498e1e578",
          "sequencer": "006A3C851F365153ED"
        }
      }
    }
  ]
}

🧠 Core Schema Extraction Points for Developers:​

• eventSource & eventName: Identifies the telemetry provider (aws:s3) and the explicit action state code (ObjectCreated:Put) that kicked off the invocation loop.
• s3.bucket.name & s3.object.key: The exact path coordinates (demo-s3-event-rendy and coffee.jpg). Your core code logic uses these variables to execute an upstream s3.get_object() call back to S3 to pull the file content chunks into memory for heavy processing.

📊 Operational Telemetry Pipeline Notation​

The asynchronous event processing metrics across your decoupling limits map according to these clear expressions:

$\text{Storage State Mutation} = \text{S3 Upload}(\text{coffee.jpeg}) \longrightarrow \text{Emit S3 Record Payload} \xrightarrow{\text{Resource-Based Policy}} \text{Lambda Invocation}$

$\text{Downstream Processing Flow} = \text{Extract}(\text{bucket.name} \;\land\; \text{object.key}) \longrightarrow \text{Execute: } \text{s3:GetObject} \implies \text{Compute In-Memory Logic}$