OpenTelemetry Operator

Kubernetes operator for managing OpenTelemetry Collectors and auto-instrumentation.

The OpenTelemetry Operator manages the lifecycle of OpenTelemetryCollector and Instrumentation custom resources. It deploys and configures OpenTelemetry Collector instances in various modes (DaemonSet, Deployment, Sidecar) and handles auto-instrumentation injection for supported languages. It is used in this cluster to centralize observability pipeline configuration, collecting traces, metrics, and logs from workloads and forwarding them to backends.

Alternatives considered

Self Hosted

Tool	Open Source	Full Features	Notes
Grafana Alloy	Yes	Yes	OTel-compatible collector; Grafana-native
Vector	Yes	Yes	High-performance data pipeline; no k8s operator
Fluent Operator	Yes	Yes	Manages Fluentd/Fluent Bit via CRDs

Installation

Architecture

HelmRelease otel-operator in namespace otel-operator, chart opentelemetry-operator version 0.109.0 from https://open-telemetry.github.io/opentelemetry-helm-charts. Admission webhooks enabled with cert-manager integration. Default collector image: otel/opentelemetry-collector-k8s. ClusterRole created for cluster-wide access. Installs OpenTelemetryCollector and Instrumentation CRDs.

Security

Manager securityContext: allowPrivilegeEscalation: false, capabilities: drop: ALL, readOnlyRootFilesystem: true, seccompProfile: RuntimeDefault. Pod securityContext: runAsNonRoot: true, runAsUser/Group/fsGroup: 65532. Admission webhooks use cert-manager for TLS. RBAC is cluster-wide (operator manages Collector pods in any namespace).

Updates

Managed by Renovate. Chart version is semver-pinned (0.109.0).

Administration

Usage

Operators deploy OpenTelemetryCollector resources to create collector instances in Deployment or DaemonSet mode. Collectors are configured with pipelines (receivers → processors → exporters) to route telemetry to Prometheus, Loki, Tempo, or other backends. Instrumentation resources enable automatic trace injection into pods via sidecar or init-container patterns without changing application code.

Cluster-specific deviations from the above live in the per-cluster README — see k8s/infrastructure/talos/controllers/otel-operator/README.md.

Cluster Deployment

OpenTelemetry Operator — Talos cluster

Cluster-specific notes only. General product info, "why we use it", and alternatives live in docusaurus/docs/platform/otel-operator.mdx.

Deviations from defaults

Defaults live in docusaurus/docs/platform/otel-operator.mdx — document anything this cluster does differently here, with a one-line reason.

Kubernetes Metadata

• HelmRelease: opentelemetry-operator@0.114.1
• HelmRepo: otel-operator (https://open-telemetry.github.io/opentelemetry-helm-charts)

Rendered manifests (kustomize build)

apiVersion: v1
data:
  values.yaml: |
    manager:
      collectorImage:
        repository: otel/opentelemetry-collector-k8s

      securityContext:
        allowPrivilegeEscalation: false
        capabilities:
          drop: ["ALL"]
        readOnlyRootFilesystem: true
        seccompProfile:
          type: RuntimeDefault

      resources:
        limits:
          cpu: 200m
          memory: 256Mi
        requests:
          cpu: 100m
          memory: 128Mi

    admissionWebhooks:
      create: true
      certManager:
        enabled: true

    clusterRole:
      create: true

    securityContext:
      runAsGroup: 65532
      runAsNonRoot: true
      runAsUser: 65532
      fsGroup: 65532
kind: ConfigMap
metadata:
  name: otel-operator-values-d84g974t45
  namespace: otel-operator