Observability
Access Decision Service (ADS) exposes three complementary observability signals that give you visibility into system health, request behavior, and security events:
- Metrics: numeric time-series data for tracking rates, errors, and latency of authorization requests.
- Distributed traces: end-to-end traces of individual requests, including spans for each attribute connector call.
- Audit logs: structured records of every authorization decision and administrative event, used for security and compliance.
For Azure-hosted deployments, ADS also supports Azure Monitor Application Insights as a unified backend for both metrics and traces.
When to use each tool
| Goal | Tool |
|---|---|
| Monitor request rates and latency | Prometheus or InfluxDB |
| Trace individual requests end-to-end | Jaeger / Zipkin (via OpenTelemetry) |
| Log who accessed what | Audit logging |
| Azure-hosted monitoring | Azure Application Insights |
In this section
- Metrics: enable and configure Prometheus or InfluxDB metrics backends.
- Distributed tracing: run ADS with an OpenTelemetry (OTel)-compatible Java agent.
- Audit logging: configure and interpret ADS audit log output.
- Azure Application Insights: integrate ADS with Azure Monitor for unified observability.