Grafana is a proud silver sponsor of this year’s KubeCon North America event, including Observability Day 2025.

OpenTelemetry Semantic Conventions define how telemetry should look, but do instrumentations actually follow them in practice? Your organization might define additional conventions - but how do you ensure instrumentations stay up to date as conventions evolve?

What if you could get a comprehensive report on the actual telemetry your application produces, checked against your conventions model?

We’re excited to share OpenTelemetry Weaver’s live check: a validation tool that listens to telemetry in real time and analyzes its compatibility against formally defined conventions. We’ll show you how to run live checks in your CI/CD pipelines to catch instrumentation issues before they hit production - no more guessing whether your telemetry matches your model.

Beyond basic model compliance, we’ll also demonstrate how to write Rego policies to enforce organization-specific checks and best practices.

Many people attending KubeCon have spent years on both Cloud Native and Kubernetes - growing in their technical knowledge alongside the development of these concepts and technologies.

But, there are others for whom that isn’t the case. There are those who are just starting out their careers. And there are others who pivoted from industries that used older methods, tools, and technologies, which is the case for one of the presenters.

This talk aims to take those very people through what it’s like to dive head-first into Kubernetes and Cloud-Native in the modern day through the eyes of an engineer who did exactly that just recently – learning it in its mature form instead of iteratively along with it. It discusses the challenges and pitfalls that can be overcome or avoided.

Additionally, it discusses what’s changed in those years that has made things easier, from the alternate perspective of an engineer who has been using Kubernetes for many years.

Prometheus and OpenTelemetry are two CNCF projects that focus on observability and truly excel at their main purposes. However, it’s no secret that they started their integration journey on the wrong foot. This story is getting better over time and being done in a data-driven way!

Through CNCF’s LFX mentorship program, Victoria and Amy conducted UX Research to understand how Prometheus should handle OTel’s Resource attributes. Using quantitative and qualitative approaches, they collected opinions from co-founders of both projects, active and old maintainers, and several end-users.

By joining this talk, the audience will learn more about what is going well and what could be improved, while listening to Victoria and Amy’s educated suggestions for the project maintainers.

Imagine you have a Kubernetes cluster that’s hosting some number of services, perhaps these services are written in different programming languages, perhaps there are some databases in the cluster too. Now, imagine that this cluster is intermittently experiencing errors and it’s not easy to tell what’s going on.

In this talk we will show you how you can add detailed telemetry immediately to a problematic production environment cluster, without any changes to your existing cluster configuration or applications, with the new OpenTelemetry eBPF Instrumentation project. We’ll show you how you’ll be able to get insights into what’s wrong with your cluster or services, by leveraging on-demand distributed traces and connectivity graphs, even if it’s the first time you have heard the term OpenTelemetry.

We’ll discuss the design principles which make this technology safe to deploy in an already problematic environment, without further compromising the stability of your cluster.

Stories of time wasted and (lots of emotions). What time do we really measure when we instrument our applications and try to collect request duration times? What conclusions do we draw when the request duration reported by client requests don’t even come close to the times reported by the server? Do we blame our network, our reverse proxies or our monitoring tools when “dark matter” gaps exist in our traces?

This talk covers some subtle nuances in request duration measurement, which can often lead to completely misleading results. We dig deep into the OpenTelemetry instrumentation approaches in a few popular programming languages and correlate what the duration timings we see in our traces or logs mean.

We’ll show how using the newly donated OpenTelemetry eBPF Instrumentation for application instrumentation allows us to augment our telemetry to overcome various inaccuracies in duration measurement… and get closer to the truth.