Blog on OpenTelemetry

OpenTelemetry JS Statement on Node.js DOS Mitigation

Thu, 15 Jan 2026 16:35:52 -0500

You may have seen a recent Node.js security advisory and related coverage discussing a potential denial-of-service issue involving async_hooks. OpenTelemetry (and other APM tools) were mentioned because we rely on AsyncLocalStorage for context propagation.

To be clear: this is not a bug or vulnerability in OpenTelemetry. The issue ultimately lies in applications and frameworks that rely on unspecified stack space exhaustion behavior for availability. In Node.js versions before 24.x, AsyncLocalStorage is implemented on top of async_hooks, which - when combined with this unsafe assumption — made the edge case easier to reproduce.

The Declarative configuration journey: Why it took 5 years to ignore health check endpoints in tracing

Thu, 15 Jan 2026 09:22:36 -0300

One of the most persistent and popular feature requests for Java OpenTelemetry over the past couple of years has been the ability to efficiently drop spans for health check endpoints – or any other low-value, cost-driving endpoints. This issue was first raised in August 2020, yet a comprehensive solution remained elusive for a surprisingly long time. Why did it take us five years to address this seemingly straightforward problem? The answer lies in the fundamental principles of OpenTelemetry’s configuration system and the journey towards a more robust, flexible approach: declarative configuration.

Calling New Contributors - Help Us Improve the OpenTelemetry Onboarding Experience

Wed, 14 Jan 2026 17:39:49 -0500

Update as of 2025-12-08

A giant thank you to those of you who have volunteered to participate! We have received more interest than expected, so sign-ups are closed for now. Stay tuned to the OpenTelemetry blog for more feedback opportunities in the future.

If you’ve ever thought about contributing to OpenTelemetry but weren’t sure where to start, you’re not alone. In a recent community survey, we heard that people were excited to contribute but got stuck taking those first steps. Contributing to open source can feel daunting, and we want to change that.

Announcing 2025 OpenTelemetry Community Awards Winners

Wed, 14 Jan 2026 07:44:34 -0500

We are excited to announce the winners of the second OpenTelemetry Community Awards! These awards recognize individuals who have made a notable impact to the OpenTelemetry project over the past year, whether it’s through code, documentation, project management, outreach, adoption, or simply helping others answer technical questions on our CNCF Slack. We received many nominations from the community, and we are delighted to share the winners with you.

Let’s congratulate our 2025 OpenTelemetry Community Awards winners:

Demystifying Automatic Instrumentation: How the Magic Actually Works

Tue, 13 Jan 2026 18:04:07 -0500

Despite the rise of OpenTelemetry and eBPF, most developers don’t know what automatic instrumentation actually does under the hood. This post breaks it down—not to suggest you build your own, but to help you understand what’s going on when your tools magically “just work.”

We’ll explore five key techniques that power automatic instrumentation: monkey patching, bytecode instrumentation, compile-time instrumentation, eBPF, and language runtime APIs. Each technique leverages the unique characteristics of different programming languages and runtime environments to add observability without code changes.

Exploring OpenTelemetry Priorities for Mainframes - Insights from Survey Responses

Tue, 13 Jan 2026 18:04:07 -0500

Which OpenTelemetry features are considered most important by users for enhancing the observability of mainframes? Earlier this year, the OpenTelemetry on Mainframes Special Interest Group (SIG) and the Open Mainframe Project conducted a survey to address this question. This blog presents a detailed overview of the survey’s findings.

Background and purpose

The OpenTelemetry project aims to enable an effective observability by making high-quality, portable telemetry available from any source to any target. The project currently hosts 90 repositories on GitHub covering specifications and implementation. When the OpenTelemetry on Mainframes SIG was formed, it gave itself the mission to enable the most important OpenTelemetry components for the mainframe and focus on the three key areas: Semantic Conventions, programming language SDKs and enhancements of the OpenTelemetry Collector. Considering the broad scope of the OpenTelemetry project and the sophisticated architecture of the mainframe, it soon became evident that a thorough understanding of user priorities is essential for making the best use of the OpenTelemetry features on the mainframe. With the results from the survey available now, the SIG will prioritize and implement targeted activities to accelerate the adoption of OpenTelemetry on the mainframe platform.

Demystifying OpenTelemetry: Why You Shouldn’t Fear Observability in Traditional Environments

Tue, 13 Jan 2026 20:49:35 +0100

For decades, traditional technology environments, ranging from on-premises data centers to legacy applications and industrial control systems, have powered the core of many organizations. These systems are battle-tested and deeply woven into business operations, but they also present unique challenges when it comes to modernizing IT practices, especially observability.

Challenges of implementing observability in traditional environments:

Noisy, unstructured logs make it hard to extract meaningful information.
Siloed monitoring data across different tools or systems leads to fragmented visibility.
Limited instrumentation in legacy apps and systems hinders collection of modern metrics and traces.
Teams are often concerned about the potential performance impact from adding new observability tooling.
Bridging legacy protocols or hardware with modern platforms can be difficult to integrate.

To make this practical, let’s follow a fictional manufacturing company with a busy production line. Here, a fleet of robotic arms equipped with sensors reports operational data via MQTT to a central broker. A legacy application logs production events and errors to disk, while a collection of SQL Servers and Windows machines support production, analytics, and inventory. Sound familiar? This is the reality for many organizations trying to bridge the old and new worlds.

OpenTelemetry.io 2025 review

Fri, 09 Jan 2026 17:52:18 +0100

As 2025 has come to an end, we’re taking a moment to look back at everything the community accomplished across the website, documentation, and localization efforts. The year was another exciting chapter for OpenTelemetry.io, and we are thrilled to share some of the highlights with you.

Highlights of 2025

Where 2024 introduced the foundation of multilingual documentation, 2025 is when localization became a core pillar of OpenTelemetry.io. This means more contributors, more translations, more supported languages, and more visibility for localized content.

What 10,000 Slack Messages Reveal About OpenTelemetry Adoption Challenges

Wed, 07 Jan 2026 18:47:50 +0100

The OpenTelemetry community has grown tremendously over the past few years, and with that growth comes valuable insights hidden in our community conversations. We analyzed nearly 10,000 messages from the #otel-collector and #opentelemetry channels on CNCF Slack spanning from May 2019 to December 2025 to understand what challenges users face most often, which components generate the most discussion, and where the community might need additional documentation or tooling improvements.

Announcing the 2025 OpenTelemetry Governance Committee Election Candidates

Mon, 15 Dec 2025 14:03:07 -0300

The OpenTelemetry Election Committee is excited to announce the final list of candidates for the upcoming 2025 OpenTelemetry Governance Committee Election!

If you are an eligible voter, you’ll have the opportunity to continue shaping the future of OpenTelemetry by casting your vote between 27 October 2025 00:00 UTC and 29 October 2025 end of day, Anywhere on Earth (30 October 2025 12:00 UTC). You’ll be selecting your preferred candidates to fill the five available seats in this year’s election.

New Configuration Schema release candidate

Fri, 12 Dec 2025 14:58:14 -0800

The latest release of the OpenTelemetry Configuration Schema is out this week. It brings the Configuration working group one step closer to completion after 3 years of effort, and the schema one step closer to being marked stable. We might be optimists here, but we think this may be the last release candidate before a stable release is done.

Many of the changes in the repositories since the previous release involved improvements to the tooling, to make the schema more consistent, and reduce the chances for new inconsistencies to be introduced in the future. There are other changes impacting both end users, contributors to the configuration schema, and implementers of the schema.

Deprecating Zipkin Exporter

Mon, 01 Dec 2025 08:40:40 -0800

The OpenTelemetry project is deprecating the Zipkin exporter specification in favor of Zipkin’s OTLP ingestion support.

Thank you to all Zipkin contributors for helping OpenTelemetry reach this milestone!

After analyzing usage patterns across language ecosystems, we’ve observed that the community has strongly gravitated toward OTLP, with Zipkin exporters seeing limited adoption — in several languages, even less than the already-deprecated Jaeger exporter. Combined with minimal user engagement on related issues and the availability of alternatives, we believe this is the right time to sunset Zipkin exporters in OTel SDKs.

Announcing Support for Complex Attribute Types in OTel

Wed, 26 Nov 2025 18:00:54 -0300

It’s common to use simple key-value properties as attributes in telemetry. Most telemetry backends are optimized for this pattern, making it efficient to store, index, and query data.

OpenTelemetry is designed with this in mind. Semantic conventions and instrumentations aim to provide useful attributes that can be easily filtered and aggregated.

But what happens when the data itself is complex? OpenTelemetry also strives to capture observability for real-world systems, libraries, and applications whose observable properties are sometimes complex.

Is the OTCA Exam Right for You? Insights for Both Newcomers and Advanced Users

Thu, 20 Nov 2025 20:06:54 -0500

In the IT industry, certifications often generate debate – some regard them as essential career milestones, while others question their practical value. While OpenTelemetry is getting widely adopted, not everyone is aware that there is a dedicated certification exam available for OpenTelemetry. The OpenTelemetry Certified Associate (OTCA) exam from the Linux Foundation is a credential designed to validate foundational knowledge and best practices in observability with OpenTelemetry, and its value extends to both newcomers and experienced professionals. This article outlines the structure of the OTCA exam, its relevance for individuals at different stages of their careers, and the benefits of pursuing this certification within the broader observability landscape.

Evolving OpenTelemetry's Stabilization and Release Practices

Fri, 07 Nov 2025 21:10:00 -0500

Summary

OpenTelemetry is, by any metric, one of the largest and most exciting projects in the cloud native space. Over the past five years, this community has come together to build one of the most essential observability projects in history. We’re not resting on our laurels, though. The project consistently seeks out, and listens to, feedback from a wide array of stakeholders. What we’re hearing from you is that in order to move to the next level, we need to adjust our priorities and focus on stability, reliability, and organization of project releases and artifacts like documentation and examples.

Contributing the Unroll Processor to the OpenTelemetry Collector Contrib

Fri, 07 Nov 2025 12:37:29 -0500

The idea for unrolling bundled logs inside the OpenTelemetry Collector didn’t start with a processor.

By “unrolling,” I mean taking a single log record that contains multiple logical events—for instance, a JSON array with ten log entries—and expanding it into ten separate log records, one for each event. This lets you work with individual log entries rather than bundled payloads.

When the Collector SIG first discussed the problem of how to handle logs that contain multiple logical events in a single body, like a JSON array, the initial instinct was to solve it with an OTTL (OpenTelemetry Transform Language) function inside the transform processor.

OpenTelemetry eBPF Instrumentation Marks the First Release

Mon, 03 Nov 2025 11:57:19 -0500

Following a significant collaboration between Grafana Labs, Splunk, Coralogix, Odigos and many other community members, we are thrilled to announce the first alpha release of OpenTelemetry eBPF Instrumentation, or OBI for short.

This event marks a significant milestone after the project, originally Grafana Beyla, was donated earlier this year by Grafana Labs. The development of eBPF instrumentation has significantly sped up after the project became managed under the OpenTelemetry umbrella. Many new protocols have been added, quality has improved - especially when deploying at scale, and tests are running 10 times faster. It’s a true testament to the value of the OpenTelemetry community.

Announcing the 2025 Governance Committee Election Results

Thu, 30 Oct 2025 16:12:35 +0000

We are delighted to announce the results of the 2025 OpenTelemetry Governance Committee Election!

The OpenTelemetry project continues to bring together an exceptionally vibrant community, who are shaping the future of observability with their contributions. This year, active contributors have shown their commitment again by participating in our election process, and they’ve done it in great numbers.

702 contributors were eligible to vote this year, and 273 of them cast their ballot. With a participation rate of 39%, it is in line with last year’s participation rate, where 299 ballots were cast with 701 eligible voters.

OpenTelemetry Sampling update

Thu, 16 Oct 2025 08:56:28 +0200

Introduction

OpenTelemetry published version 1.0 of its Tracing specification over four years ago, and the same year W3C TraceContext Level 1 was published with W3C Recommendation status. We as a community and we the observability industry had two new standards for distributed tracing. Of course, we weren’t finished.

Sampling is a major topic of the Tracing SDK specification, and the original specification included a set of built-in Samplers, AlwaysOn, AlwaysOff, ParentBased, and TraceIdRatioBased, along with an interface allowing new samplers to be implemented, primarily Jaeger Remote.

OTel Unplugged EU at FOSDEM 2026

Fri, 10 Oct 2025 06:53:18 -0400

OpenTelemetry is coming to FOSDEM!

When we put out our community survey, you requested more meetups and we have heard you! We’re happy to announce that we are bringing back OTel Unplugged, the OpenTelemetry unconference that we first ran in 2022.

What is OTel Unplugged?

OTel Unplugged is an OpenTelemetry unconference and project roadmapping session. An unconference is like a conference, only instead of a speaker track, we have a series of breakout sessions where attendees get to pick the topics. If you are an end user, this is a great opportunity to connect with maintainers and other users to get your questions answered and give feedback to the project. If you are a maintainer or contributor, the is your chance to connect with your users face to face!