Serverless AWS Lambda Development · レッスン

AWS X-Rayによる分散トレーシング

AWS X-Rayを使って分散トレーシングを実装・解釈し、サーバーレスアーキテクチャ内のリクエストの流れを可視化して、パフォーマンスのボトルネックを特定します。

「AWS X-Rayによる分散トレーシング」はCoddyKit上の無料Serverless AWS Lambda Developmentレッスンです。これはレッスン3/4です。下記で完全なレッスンを無料で読むことができます。その後、ブラウザ内の組み込みコードエディタと24時間対応のAIチューターでハンズオン演習できます。これはServerless AWS Lambda Development学習パスの一部であり、ウェブとCoddyKitアプリ全体で進捗が同期されます。 Serverless AWS Lambda Developmentコースには全4レッスンが含まれています。

このレッスンの一部はまだ翻訳されておらず、英語で表示されています。

Tracing Serverless Requests

In serverless applications, a single user request often travels through many services, like an API Gateway, multiple Lambda functions, and databases. Understanding this journey can be complex!

This is where distributed tracing comes in. It's a technique to track a request as it flows across these different services, providing a complete end-to-end view.

Meet AWS X-Ray

AWS X-Ray is a service that helps developers analyze and debug distributed applications built using microservices. It provides an end-to-end view of requests as they travel through your application.

With X-Ray, you can:

Visualize service interactions.
Identify performance bottlenecks.
Debug errors across distributed systems.

X-Ray Segments Explained

The basic unit of data in X-Ray is a segment. Each service (like a Lambda function or an API Gateway) that's part of your application sends its own segment to X-Ray.

A segment contains information about the work done by that service, including:

The service's name and ID.
Details about the incoming request.
Timing information (start and end times).
Any errors or faults that occurred.

Deeper Dive: Subsegments

Within a service's segment, you can create subsegments for more detailed tracing. Subsegments represent discrete units of work within that service.

For example, a Lambda function's segment might contain subsegments for:

Calls to a database (like DynamoDB).
HTTP requests to an external API.
Specific business logic within your code.

They help pinpoint exactly where time is being spent inside a service.

Traces and The Service Map

All segments and subsegments generated by a single request are grouped together to form a trace. This trace provides a complete, end-to-end view of the request's journey.

X-Ray then uses these traces to generate a service map. This is a visual representation of your application's components and the connections between them, showing latency and error rates.

Enable X-Ray for Lambda

To start tracing your Lambda functions, you first need to enable X-Ray tracing for them. This can be done easily through the AWS Management Console or via Infrastructure as Code (like AWS SAM or CloudFormation).

In the Console:

Go to your Lambda function.
Under 'Configuration', select 'Monitoring and operations tools'.
Edit and enable 'Active tracing' for AWS X-Ray.

This sets up Lambda to send basic function invocation data to X-Ray.

Instrument Python Lambda Code

While enabling X-Ray captures basic data, for deeper insights (like custom subsegments or tracing non-AWS SDK calls), you need to instrument your code using the X-Ray SDK.

Here's a Python example:

from aws_xray_sdk.core import xray_recorder
from aws_xray_sdk.core import patch_all

# Patch all AWS SDK calls automatically
patch_all()

def lambda_handler(event, context):
    # Start a custom subsegment for specific logic
    with xray_recorder.in_subsegment('## my_custom_logic'):
        print("Executing custom business logic...")
        # Simulate some work
        import time
        time.sleep(0.05)

    # Any AWS SDK calls after patch_all() will be automatically traced.
    # For example, a DynamoDB put_item() call.

    return {
        'statusCode': 200,
        'body': 'Hello from Lambda with X-Ray!'
    }

# For local testing (not typically deployed with Lambda)
if __name__ == '__main__':
    print(lambda_handler({}, {}))

Instrument Node.js Lambda Code

Similarly, for Node.js Lambda functions, you use the aws-xray-sdk package to instrument your code. This allows you to capture custom subsegments and ensure AWS SDK calls are traced.

Here's a Node.js example:

const AWSXRay = require('aws-xray-sdk');

// Patch all AWS SDK calls automatically.
// This ensures calls to other AWS services are traced.
AWSXRay.captureAWS(require('aws-sdk'));

exports.handler = async (event, context) => {
    // Get the current segment created by Lambda's X-Ray integration.
    const segment = AWSXRay.getSegment();

    // Create a custom subsegment to trace specific logic.
    if (segment) { // Ensure segment exists for safe local execution
        const customSubsegment = segment.addNewSubsegment('## MyCustomLogic');
        try {
            console.log("Executing custom business logic...");
            await new Promise(resolve => setTimeout(resolve, 50)); // Simulate work
        } finally {
            customSubsegment.close();
        }
    } else {
        console.log("X-Ray segment not found, executing without tracing.");
        await new Promise(resolve => setTimeout(resolve, 50)); // Simulate work
    }

    return {
        statusCode: 200,
        body: JSON.stringify('Hello from Lambda with X-Ray!'),
    };
};

// For local execution, you can simulate a call to the handler.
if (require.main === module) {
    console.log("Running handler locally...");
    exports.handler({}, {}).then(result => console.log(result));
}

Reading the Service Map

Once X-Ray collects trace data, you can view the service map in the X-Ray console. This map visually represents your application's components as nodes and their interactions as edges.

Look for:

Nodes: Represent services (Lambda, API Gateway, DynamoDB).
Edges: Show connections and data flow between services.
Colors: Indicate service health (green for healthy, red for errors).
Latency: Visual cues and metrics on edges show request duration.

Pinpointing Bottlenecks

The real power of X-Ray is in identifying performance issues. By examining a trace's timeline, you can see how long each segment and subsegment took to complete.

This helps you:

Identify slow services: Which service is taking the most time?
Find inefficient code: Which subsegment within a function is causing delays?
Detect external dependencies: Is a third-party API call or database query slowing things down?

X-Ray Tracing Check

Let's test your understanding of AWS X-Ray's capabilities.

Tracing the Path Forward

Congratulations! You've learned how AWS X-Ray provides invaluable visibility into your serverless applications.

By enabling X-Ray, instrumenting your code, and interpreting the service map and trace timelines, you can effectively:

Understand complex request flows.
Diagnose latency issues and errors.
Optimize your application's performance.

Embrace X-Ray to build more robust and performant serverless systems!

よくある質問

「AWS X-Rayによる分散トレーシング」レッスンは無料ですか？

はい。「AWS X-Rayによる分散トレーシング」の完全なテキストはこのウェブで無料で読めます。インタラクティブに演習し（組み込みコードエディタと24時間対応のAIチューター）、Serverless AWS Lambda Developmentコースの残りをアンロックするには、CoddyKit PROにアップグレードしてください。 Serverless AWS Lambda Developmentコースには全4レッスンが含まれています。

「AWS X-Rayによる分散トレーシング」で何を学びますか？

AWS X-Rayを使って分散トレーシングを実装・解釈し、サーバーレスアーキテクチャ内のリクエストの流れを可視化して、パフォーマンスのボトルネックを特定します。ブラウザで直接実行するハンズオンコードでServerless AWS Lambda Developmentを演習し、24時間対応のAIチューターがレッスンを進める中での質問に答えます。

Serverless AWS Lambda Developmentを始めるのに経験は必要ですか？

事前経験は必要ありません。CoddyKitのServerless AWS Lambda Developmentは初級者から上級者向けに構成されているため、ここから始めるか最初から始めて、自分のペースで進むことができます。これはレッスン3/4です。

「AWS X-Rayによる分散トレーシング」レッスンにはどのくらい時間がかかりますか？

ほとんどのCoddyKitレッスンは約5～10分かかります。各レッスンはコンパクトでインタラクティブなので、着実に進歩し、ウェブとアプリ全体で正確に前回の場所から再開できます。

このServerless AWS Lambda Developmentレッスンでコードを書いて実行できますか？

はい。すべてのServerless AWS Lambda Developmentレッスンに組み込みコードエディタが含まれているため、ブラウザでリアルコードを書いて実行し、即座のAIフィードバックを取得できます。ローカル設定は不要です。

このコースのすべてのレッスン

← Serverless AWS Lambda Developmentに戻る