Serverless introduces a number of challenges to existing tools for observability, we need to adapt our practices to fit this new paradigm. In this talk we will see how you can implement log aggregation, tracing and correlation IDs for a serverless application.
In this talk we’ll introduce the challenges the serverless paradigm introduces to existing tools we use for observability:
- nowhere to install monitoring agents/daemons
- no background processing
- higher concurrency to monitoring systems
- lots more asynchronous invocations
We’ll then discuss how you can build observability into a serverless application, whilst touching on:
- how to implement log aggregation
- the need for structured logging and sampling debug logs in production
- how to capture and forward correlation IDs through both synchronous (API Gateway) as well as asynchronous (SNS, Kinesis, S3, etc.) invocation events
- how to capture performance traces to help debug performance issues
Getting started with AWS Lambda is easy, but understanding the operational challenges and tradeoffs is a much more nuanced topic, and one that you need to grasp before you can be confident in operating a serverless architecture responsibly. In this workshop we will take you beyond the basics and teach you what you need to know to make your serverless application production ready!
In this workshop we will cover:
- AWS Lambda 101
- Serverless framework 101
- Writing integration and acceptance tests for AWS Lambda functions
- CI/CD with CodePipeline and CodeBuild
- Managing configurations with SSM Parameter Store
- Per-function IAM roles
- Structured logging
- Sample logging
- X-Ray integration
- Capturing and forwarding correlation IDs
