Monitoring¶

Exposition API¶

L API expose:

GET /metrics (Prometheus via OpenTelemetry)
GET /health/live
GET /health/ready

flowchart LR
  API[API AdvancedDevSample] --> MET[/metrics/]
  API --> LIVE[/health/live/]
  API --> READY[/health/ready/]

Instrumentation¶

OpenTelemetry configure:

traces: ASP.NET Core + HttpClient
metriques: ASP.NET Core + HttpClient + exporter Prometheus
export OTLP optionnel via OTEL_EXPORTER_OTLP_ENDPOINT ou OpenTelemetry:Otlp:Endpoint

flowchart LR
  REQ[Requetes HTTP] --> OTel[Kit OpenTelemetry]
  OTel --> TR[Traces]
  OTel --> M[Metriques]
  TR --> OTLP[Endpoint OTLP optionnel]
  M --> PROM[/metrics pour Prometheus]

Stack locale¶

Fichiers:

docker-compose.yml
monitoring/prometheus/prometheus.yml
monitoring/prometheus/alert.rules.yml
monitoring/grafana/provisioning/**
monitoring/grafana/dashboards/api-observability.json

Lancement:

docker compose --profile monitoring up -d

Acces:

Prometheus: http://localhost:9090
Grafana: http://localhost:3000 (admin / admin)

Alertes par defaut¶

AdvancedDevSampleHigh5xxRatio
AdvancedDevSampleHighLatencyP95
AdvancedDevSampleNoTraffic

flowchart TD
  P[Collecte Prometheus] --> R[Regles d enregistrement et alertes]
  R --> A1[High5xxRatio]
  R --> A2[HighLatencyP95]
  R --> A3[NoTraffic]
  A1 --> N[Canal d alerte]
  A2 --> N
  A3 --> N

Verification rapide¶

lancer la stack monitoring
verifier http://localhost:5069/metrics
verifier la cible API en UP dans Prometheus
ouvrir le dashboard Grafana AdvancedDevSample API Observability

flowchart LR
  S[Demarrer le profil monitoring] --> M[Verifier /metrics]
  M --> T[Cible API UP dans Prometheus]
  T --> G[Ouvrir le dashboard Grafana]