OpenTelemetry

OpenTelemetry native. From day one. No lock-in.

Every OTel-compatible SDK, collector and agent works with Obseria immediately. Open standards for data, queries and dashboards
Obseria apps and capabilities: observability, APM, infrastructure and integrations
Obseria home dashboard with service health, open problems, SLO compliance and ingestion
Obseria trace explorer with span list and span attributes
Benefits

Open standards, full control

Obseria builds on OpenTelemetry, Prometheus and Perses instead of proprietary formats.

OpenTelemetry native

OTLP for logs, metrics, traces and profiles. Semantic conventions are fully supported, including GenAI and Kubernetes.

PromQL, SQL and OQL

Keep using existing PromQL queries and alert rules, use SQL for ad-hoc analysis or OQL for cross-signal questions.

No lock-in

Export data at any time in OTLP format. Version dashboards as Perses code. No proprietary formats, no lost context.

0 proprietary agents required

Fjord Software uses OTel SDKs and the collector exclusively.

Read the Fjord story

100% of PromQL alerts migrated

Deltaforge migrated 1,200 Prometheus rules without changes.

Read the Deltaforge story

3 days to migrate from a proprietary backend

Kestrel Mobility switched by changing one collector exporter.

Read the Kestrel story

40+ contributions to OpenTelemetry projects

Obseria engineers contribute to the collector, semantic conventions and SDKs.

Engineering blog

Empower your business with AI-driven observability

Harness trusted answers instead of alert floods, for measurable business outcomes at scale

Explore the Obseria approach
How it works

From signal to answer in three steps

  1. Step 1

    Point your exporter

    Send OTLP from existing SDKs and collectors to Obseria, or use Prometheus remote write.

  2. Step 2

    Keep your conventions

    Semantic conventions, resource attributes and PromQL rules work unchanged.

  3. Step 3

    Stay portable

    Export data in OTLP at any time and version dashboards as Perses code.

What's included

  • OTLP for logs, metrics, traces and profiles
  • Full semantic conventions incl. GenAI
  • PromQL alerts and Alertmanager import
  • Perses dashboards as code
  • Validated collector configurations
  • Export at any time, no lock-in
Use cases

OpenTelemetry use cases with Obseria

AAcme PlatformProAI AgentHomeDashboardsCatalogSearchMonitoringInfrastructureServicesTracesLogsExperienceSLOsIncidentsSecurityWorkflowsBusinessData connectionsFind what you need⌘ KData connections · OTLPLast 1 hourEnvironmentAll ⌄ServiceAll ⌄StatusAll ⌄SeverityAll ⌄RefreshData connections · OpenTelemetry pipelineAdd sourceOTel SDKsJava · .NET · Go · PythonPrometheusremote write · 1.2M seriesKubernetesoperator · 12 clustersCloud APIsAWS · Azure · GCPObseria PipelineFilter−62% debugMask PII14 fieldsEnrichk8s + traceSampletail-basedRoute3 bucketsLogs3.3B / monthMetrics12.9B pointsTraces1.2B spansProfilescontinuousThis monthIngested2.4M/sStored after filtering38%Saving$14,820Dropped errors0

One collector, all signals

Obseria provides validated collector configurations for Kubernetes, VMs and serverless. Processors for sampling, enrichment and masking run before export.

Documentation
Integrations

Works with your stack

Connect the technologies you already run. 700+ integrations, discovered and put into context automatically.
Browse all integrations

Leading organizations that run on OpenTelemetry

  • Bosch
  • Ströer
  • Caritas international
  • Valensina
  • Deutsche Aidshilfe
  • SOFAR
  • Hitchcock
  • Schmid Alarm
  • HiScout
  • vi2vi
  • Kienzle Dentaltechnik
  • Flowcon Unternehmensberatung
  • Marissa Resort
  • Hansjakob Realschule Freiburg

OpenTelemetry FAQ

OpenTelemetry is a Cloud Native Computing Foundation (CNCF) project and the open industry standard for generating, collecting and exporting telemetry data: traces, metrics, logs and profiles.

No. The agent is optional and adds automatic instrumentation, profiling and security to OpenTelemetry. Pure OTel setups are fully supported.

Yes. PromQL queries and alert rules work unchanged. Alertmanager configurations can be imported.

At any time in OTLP format via API or as a batch export to your object storage. Dashboards are available in Perses format.

Ready for observability that delivers answers?

Take the first steps toward reliable operations