API observability fundamentals

The opening keynote of the API observability fundamentals programme, with Budha Bhattacharya and Sonja Chevre.

A Panel discussion that takes a deeper look at observability featuring a Navish Bahl, Abby Bangser, Andreas Grabner, and Ruchir Jha

Budha and Sonja wrap-up the API observability fundamentals programme.

Developers and engineers, often not in decision-making roles, regularly face challenges in troubleshooting, diagnosing, and resolving issues in environments that lack observability practices. We feel the direct impact of inadequate observability on system reliability and performance. In this session, we will explore the critical role of observability in modern software development and delve into strategies for effectively advocating its adoption within our teams. Drawing from personal experiences, I will share insights on how attendees can make a compelling case for the necessity of observability and champion the adoption of observability practices within their teams.

With the growing complexity of modern software systems, having good observability is essential for developers to diagnose and fix issues, improve performance, and enhance the overall user experience.

OpenTelemetry is a powerful and flexible open-source observability framework that enables developers to capture, generate, collect, and export telemetry data such as traces, metrics, and logs from their applications.

However, getting started with OpenTelemetry can be daunting for developers who are new to the framework. In this talk, Juliano will share tips and best practices for overcoming the initial challenges of implementing OpenTelemetry, using the OpenTelemetry Demo as a guide. The demo provides a comprehensive overview of how to use OpenTelemetry with almost all supported programming languages.

Learn how to generate traces, metrics and logs using both auto and manual instrumentation, how and why to use the Collector, and how to send the collected data to open-source tools.

Telemetry without context is just data. In this talk, we look at the importance of contextual metadata for telemetry, how OpenTelemetry resources solve that problem, and what tooling is available to configure resource metadata in OpenTelemetry.

In this presentation, we will start with a traditional Playwright test against a distributed service running under the Tyk gateway. Tests will still return as passing even though downstream services are breaking. We’ll explain how to incorporate modern observability techniques to include distributed traces and trace-based testing in your Playwright tests.

And how you can break the chain of blame, properly identifying what part of the system broke and assigning the issue to the proper team.

We will present how to use distributed traces to visualise a defined code path and have it be part of a Playwright end-to-end test.

This talk will be helpful for developers working with Cloud Native distributed systems helping them test all possibilities and deploy code safely and reliably.

A repo will be provided after the presentation that you can clone. It will allow you to spin up and play with the NodeJS-based demo app running behind the Tyk gateway with Tracetest-enabled Playwright tests.

Join us for an immersive session where we address the specific challenges of API management and troubleshooting. No slides, just raw storytelling and practical demonstrations. I'll guide you through the process of utilising different APIs, encountering intentional failures like rate limiting, quotas, and even server shutdowns along the way. As we explore deeply into the domain of observability, leveraging its insights to troubleshoot and resolve potential issues in real-time. Join me for an engaging session that promises a hands-on approach to mastering API Observability.

Defining the right configuration for pinging your API service, with the right level of testing and frequency, is a tricky balancing act. We'll look at how to balance cost and signal clarity with a need for site reliability.

• What is a heartbeat check?
• Defining your SLA
• Calculating your config
• More details on the configuration
• Tool options

Henrik Rexed dives into how to efficiently autoscale your workload in K8s using observability using real work examples.

Join us as we explore how teams can streamline troubleshooting and root cause analysis using OpenTelemetry, Pyroscope, Grafana and Tempo. We'll delve into best practices for utilizing RED metrics, demonstrating how it guides effective monitoring and prioritization of performance issues. Starting with a refresher on Opentelemetry, we’ll progress through hands-on examples showcasing the power of TraceQL for querying distributed traces in Tempo, enabling precise analysis, and dive deep into code-level performance insights with Pyroscope. Don't miss this opportunity to optimize your troubleshooting workflows and drive continuous improvement in system performance and collaboration.