Code Red Newsletter #4

Hi there,

Welcome back to your bi-weekly hit of observability goodness, where we explore telemetry and all the interesting things you can do with it.

Every week we have a topic. And this week’s topic is the fine balance between too much data and not enough context.

Disclaimer: We write this newsletter with an eye firmly on what is happening in the community. However, this particular installment might come across uncharacteristically Dash0-centric.

Because, you see, the thing is: we have released a lot of really cool stuff in the past couple weeks that enables you to walk with confidence and grace that fine balance between too much data and not enough context.

In Focus: Use the right telemetry, the right way

When we started Dash0, we had some very precise ideas about what an OpenTelemetry-native observability tool should look like. And, while it is just one year ago (almost to the day), in our perspective it is aging like a fine wine.

In the past few weeks, we made strides towards fulfilling our original view. Specifically, we have given you:

1. An automated way of getting insights about issues and outliers from your telemetry at large
2. Streamlined, effective means of curbing unnecessary telemetry

Let’s dive right into it.

Triage: The Quickest Way to Finding the Needle in the Haystack

In the past few years, collecting telemetry has become so much easier. And since we can collect tons of telemetry, well, we did, and now we are all amassing great haystacks of data. Trying to find patterns and correlations in such large amounts of data is a daunting prospect for the most experienced of operators.

Enter Dash0 Triage - a magnet that lets you pull needles out of your telemetry haystack in fractions of seconds and none of the haystack diving and sifting. You tell Dash0 what you are interested in by selecting areas on the Dash0 Tracing Heatmap or our RED Charts (more on this below). Dash0 then tells you what is different about it, be it a rogue service ID, a misbehaving request, or that one product SKU that just won’t behave. No more endless clicking, just instant insights.

Triage is in beta for Tracing, with Logging already in the works and more to follow, but it’s already saving engineers from telemetry-induced existential dread. Our own SREs cannot fathom going back to not having it.

Want to see it in action? Check it out here.

Spotlight: RED Charts in Tracing Explorer

Dash0’s Tracing Heatmap is a very powerful visualization optimized to point out outliers in terms of latency or error rate (or both). But sometimes, you just want to break down what you see in the three dimensions of observability: Requests, Errors, and Duration (RED).

The Dash0 RED Charts do just that: they are an alternative visualization to the Tracing Heat map that puts RED metrics front and center, making it easy to spot trends, detect anomalies, and troubleshoot faster. Whether you’re tracking error spikes or latency issues, RED Charts help you cut through the noise and focus on what matters.

Read more about it here.

Cut Through the Noise: Advanced Grouping in Dash0

If you have used Dash0, you are familiar with the powerful filtering features. And now, they are complemented by grouping: with one click, you can group all your logs or spans across specific attribute keys, and can instantly spot patterns, identify anomalies, and drill down.

Observability shouldn’t feel like searching for a pixel in a 4K image.

See it in action here.

Polish that telemetry to a shine: Automatic upgrade of semantic conventions

This bit we already talked about in the last installment, but it fits the theme so well that we have to bring it up again: Dash0’s automatic schema migrations ensure your telemetry follows OpenTelemetry’s latest conventions - so you can focus on insights, not syntax.

Read more here.

Spam Filters: The Point-and-Click Way to Stay Within Your Observability Budget

Telemetry is great - until your budget starts looking like a cryptominer’s power bill. OpenTelemetry and Prometheus have made collecting data easy, but filtering out what you don’t need? That’s still toilsome and error prone.

Enter Dash0 Spam Filters: a point-and-click way to drop useless logs, metrics, and spans before they hit your bill. You use the full power of Dash0 filtering, and with one click (well, two, if we are splitting hair: there is a confirmation dialog), you drop the telemetry you don’t care for. What you drop, you don’t pay for. And since we like simple pricing and happy users, you won’t pay for the filtering itself either.

No YAML acrobatics, no regex rituals - point-and-click to rightsizing your observability budget. You can also export the rules to YAML that you can use with the OpenTelemetry Collector or the Dash0 Operator (more on this below).

Filter out successful Kubernetes probes? Done. Drop verbose logs from staging? Easy. Your CFO will s̶t̶o̶p̶ ̶h̶a̶u̶n̶t̶i̶n̶g̶ ̶y̶o̶u̶ thank you. See how it works here.

Look, more filtering, and this time directly from your Kubernetes cluster!

Spam filters operate in the Dash0’s ingestion pipeline. But maybe you want to redact telemetry already from within your Kubernetes cluster. This way, for example, you can spare telemetry egress costs.

The Dash0 operator now allows you specify filter rules using the same syntax that the OpenTelemetry Collector and OpenTelemetry Operator do, but in a way that feels cohesive with the overall Dash0 Operator experience.

Code RED Podcast: Behind the Screens with Dash0 CTO Ben Blackmore

What happens when a seasoned engineer steps into a CTO role? Dash0’s own Ben Blackmore joins Mirko Novakovic for a deep dive into the evolving world of observability - and what it’s like to go from shipping code to leading a team of 20+ engineers.

They chat about real-world Code Red moments, the challenges of scaling AI-powered logging, and why web developers think differently (hint: “works on my machine” is a valid troubleshooting step). Plus, hear Ben’s take on why tracing and RUM should finally get along and how OpenTelemetry is shaping the future of frontend observability.

It’s candid, insightful, and full of stories that every engineer-turned-leader can relate to. Listen to the episode here.

Is OTel the Last Observability Agent You’ll Ever Install?

The Observability world has changed - for good. OpenTelemetry (OTel) isn’t just another tool; it’s rewriting the rules, making proprietary agents feel like relics of a bygone era.

With OTel, vendor lock-in is out, and switching providers is as easy as updating the configurations of an endpoint and an authentication token.

But is this all there is to the OpenTelemetry revolution? No, it goes way deeper than that: Vendors now must compete on insights, efficiency, and cost, rather than just locking you in with proprietary telemetry plumbing. Observability is no longer about how you collect data - it’s about what you can do with it.

So, is OTel the future? Read on and decide for yourself. Check it out here.

Unlocking Kubernetes Observability with the OpenTelemetry Operator

As Kubernetes scales, so does its complexity. Monitoring workloads and infrastructure shouldn’t feel like fighting a hydra - cut down one challenge, and two more appear. The OpenTelemetry Operator simplifies observability by automating telemetry collection, eliminating manual configuration headaches, and giving you full visibility without the toil.

With automated OpenTelemetry Collector deployments and seamless auto-instrumentation, you can capture logs, metrics, and traces effortlessly across your entire cluster. Whether you're integrating with Prometheus or moving towards an OTel-native approach, the OpenTelemetry Operator makes it easy to scale observability without drowning in YAML.

Read the full breakdown here.

Choice Cuts: The Best of What’s New

Why Is Multicloud Observability So Expensive?

Observability is critical for modern cloud-native systems - but why does it feel like you need a second budget just to keep your logs, metrics, and traces in check? The answer lies in data explosion, vendor pricing models, and cardinality creep.

With OpenTelemetry making data collection easier than ever, teams often collect everything, only to realize later that storage and query costs skyrocket. Vendors charge based on ingestion, retention, and query complexity, meaning your observability bill can spiral out of control even if half your data is noise.

This is exactly why smarter filtering (yes, like those handy Spam Filters we just talked about) is becoming essential. Dropping unnecessary data before it inflates your bill is the easiest way to keep observability costs in check without losing insights.

Find out more here.

OpenTelemetry Talks and Activities at KubeCon+CloudNativeCon Europe 2025

KubeCon Europe is fast approaching, and OpenTelemetry is everywhere. From hands-on ContribFest sessions to deep-dive technical talks, this is the place to connect with maintainers, learn from real-world case studies, and get involved in shaping the future of observability.

Expect key sessions on AI-powered telemetry, database observability, and tracing at scale, alongside maintainer updates and panel discussions covering the latest in OpenTelemetry adoption. Observability Day will also feature expert talks, including a session from Dash0’s Michele Mancioppi on no-touch instrumentation.

If you're interested in the intersection of observability and platform engineering, I'll be giving a keynote on Wednesday exploring how the two come together to create scalable, insight-driven platforms.

Check out the full schedule here.

See You in Two Weeks

That’s it for this installment of the Code RED Newsletter! The world of Observability is evolving fast and is full of possibilities. This is a wonderful ride, and we’re glad to have you along for it.

We’ll be back in two weeks with more deep dives, industry updates, and (let’s be honest) many other chances to talk about cool new things about OpenTelemetry and Observability at large.

Also, the Dash0 team will be onsite in London for KubeCon+CloudNativeCon, make sure to stop by our booth for a chat, demo, and a chance to win some cool prizes.

Kasper, out.