Recent changes and improvements to Dash0

Our log ingestion pipeline uses AI to understand the log structure of applications and automatically extract the severity. Previously in Beta, this capability is now available for everyone, out of the box. And at no extra charge!

Logs with known severity like INFO, WARN and ERROR are not only easier to filter in the UI, but can also be used as system health indicators when setting up dashboards and check rules.

Built on the foundation of the dash0.spans metric, dash0.span.events empowers you to query the number of events associated with your spans. Its advanced filtering and grouping capabilities give you full access to span and event attributes, such as event names (otel.span.event.name), offering a new level of insight into your trace data via PromQL.

This metric is also at the heart of the all-new Dash0 Cost Estimate dashboard, now available in the Integration Hub. By breaking down event volumes across services, the dashboard provides a clear view of how your event data contributes to overall observability costs. With this knowledge, you can make smarter, data-driven decisions to manage and optimize expenses while maintaining visibility.

Start exploring dash0.span.events today and experience a more detailed, actionable view of your trace data — helping you control costs.

As part of this update, our two thresholds annotations have been renamed to better align with other Dash0-specific extensions:

• threshold-degraded -> dash0-threshold-degraded

• threshold-critical -> dash0-threshold-critical

Additionally, we've reimagined our handling of the severity label, offering you the freedom to assign any custom value that fits your needs.

These updates reflect our commitment to providing a seamless and intuitive experience for users adopting the Dash0 operator. For new users, we recommend adopting these updated naming conventions as detailed in our documentation. Existing users managing Dash0 Check Rules with the Prometheus alert rule format can rest easy—if a migration becomes necessary, we'll reach out to provide step-by-step guidance, ensuring a seamless transition.

With the Dash0 operator, managing check rules is simpler, more powerful, and tailored to your environment.

You can learn setup instructions through the Integrations Hub that help you get started quickly with AWS, Vercel, Node.js, and much more. It even comes with ready-made dashboards that model everyday observability needs, e.g., for the OpenTelemetry collector, Vercel, GitHub Actions, and more. We plan to extend it in the future with content about alerting and more. Check it out!

With this feature, Dash0 alerting notifications can be configured to route into Alertmanager for handling alerts, leveraging its existing mechanisms for routing, deduplication, and grouping. This provides an opportunity to integrate Dash0’s alerting with workflows already built around Prometheus and its related tools, without disrupting your current setup.

This notification channel is available in Beta and joins our recently added incident.io and BetterStack integrations. Together, these features reflect our ongoing commitment to giving you more options to align Dash0 with your existing observability workflows, whether they’re centralized within Dash0, or involve other tools in your ecosystem.

We’re excited for you to try this feature and explore how Dash0 can fit seamlessly into your alerting strategy.

When looking at a chart, you really don't want to present information about errors in a neutral (or even positive) color tone. Instead, you will want the color to signal that something is off by making it red or yellow. This now works by default within Dash0!

Dash0's semantic coloring system (see our article "Why is this red?") now works for custom charts, too!

How does it work? When a time series is labeled with an attribute containing status-like information, such as log severities, span status codes, HTTP & gRPC status codes, etc., the charts automatically pick the right colors!

Learn more within our documentation on panel options.

The Dash0 operator is an open-source Kubernetes operator built on OpenTelemetry, Prometheus, Perses and other open-source projects, that provides you with an appliance-like way (one command, and all works!) of monitoring your Kubernetes clusters and the applications running on top of it.

Today we release the 0.37.0 version of the Dash0 operator, which automatically installs and configures the OpenTelemetry Java agent in your Kubernetes pods to monitor your Java applications out of the box.

Try it out and let us know what you think of it!

Dash0’s Vercel integration will automatically send all logs from your Vercel workloads to Dash0 for analysis and visualization. Instead of configuring a log drain in Vercel manually, you can just install the integration at https://vercel.com/integrations/dash0 and we do it all for you.

When the integration has been installed, logs from your Vercel projects will be sent automatically to Dash0. This is great news as the log retention within Vercel is only 3 days, while Dash0 offers 30 days of retention for logs and traces. You will also be able to analyze your Vercel logs in context with telemetry from your non-Vercel infrastructure, which makes troubleshooting so much easier.

Additionally, we have integrated the extraction of host and process data from these logs, which is also aligned with OpenTelemetry standards. This improvement significantly enhances your ability to analyze and interpret Journald log data effectively.

OpenTelemetry journald receiver, https://github.com/open-telemetry/opentelemetry-collector-contrib/tree/main/receiver/journaldreceiver

Having good resource attributes on your telemetry is paramount to understanding what is going on in your workloads. Logs, metrics and traces without resource information are basically just data without context. And yet, making sure there are correct Kubernetes resource attributes on all your telemetry is not always easy.

That v0.36 of the Dash0 operator now automatically adds k8s.namespace.name, k8s.pod.name, k8s.pod.uid, and k8s.container.name to all workloads via the Dash0 injector. Under the hood, these attributes are added to the resource attributes sent by your applications via the OTEL_RESOURCE_ATTRIBUTES environment variable. If the workload uses an OpenTelemetry SDK, this environment variable is automatically picked up, and the attributes defined in it are sent along with all the spans, metrics and log records that the application emits.

What's more, the Kubernetes Attributes Processor (which is deployed automatically by the Dash0 operator) will enrich telemetry with a host of other Kubernetes related resource attributes, like the Kubernetes namespace identifier, deployment name and unique identifier, replicaset name and unique identifier etc.

One less thing to worry about, one more thing that just works with Dash0.

Big news: Your workflow, your rules! 🎉 Personalize Dash0 like never before and keep your most-used views just a click away. Dive in and make it yours!

A common problem when creating a dashboard is forgetting to save. Painfully crafted queries and panels are lost. It's so annoying when that happens! No more with Dash0, though!

We even synchronize state across browser tabs for all those tab hoarders! 👀