Jak nasadit OpenTelemetry v Kubernetes¶

OpenTelemetry (OTel) je open-source standard pro observabilitu — traces, metriky a logy z jednoho SDK. Tady je návod, jak ho nasadit v Kubernetes.

Předpoklady¶

• Kubernetes cluster (1.26+)
• kubectl + helm
• Grafana stack (Tempo, Mimir, Loki) nebo jiný backend

Krok 1: Instalace OTel Collector¶

helm repo add open-telemetry https://open-telemetry.github.io/opentelemetry-helm-charts
helm install otel-collector open-telemetry/opentelemetry-collector \
  --namespace observability \
  --create-namespace \
  -f values.yaml

Krok 2: Konfigurace Collector¶

# values.yaml
mode: deployment
config:
  receivers:
    otlp:
      protocols:
        grpc:
          endpoint: 0.0.0.0:4317
        http:
          endpoint: 0.0.0.0:4318
  processors:
    batch:
      timeout: 5s
      send_batch_size: 1000
  exporters:
    otlp/tempo:
      endpoint: tempo.observability:4317
      tls:
        insecure: true
  service:
    pipelines:
      traces:
        receivers: [otlp]
        processors: [batch]
        exporters: [otlp/tempo]

Krok 3: Instrumentace aplikace¶

Pro Python:

from opentelemetry import trace
from opentelemetry.sdk.trace import TracerProvider
from opentelemetry.exporter.otlp.proto.grpc.trace_exporter import OTLPSpanExporter

provider = TracerProvider()
exporter = OTLPSpanExporter(endpoint="otel-collector.observability:4317")
provider.add_span_processor(BatchSpanProcessor(exporter))
trace.set_tracer_provider(provider)

Krok 4: Ověření¶

kubectl port-forward svc/grafana 3000:3000 -n observability
# Otevřete Grafana → Explore → Tempo → hledejte traces

Tipy¶

• Sampling: Pro produkci nastavte tail-based sampling (1-10% traces)
• Resource detection: Přidejte k8s resource detector pro automatické labely
• Cost: OTel je zdarma, ale backend storage stojí peníze — plánujte retenci

Závěr¶

OpenTelemetry je budoucnost observability. Jednou instrumentujete, backend měníte kdykoliv.