Skip to content

Privacera Monitoring

The Self-Managed version of Privacera supports monitoring and observability of both Privacera applications and the underlying infrastructure. The observability stack collects logs, metrics, traces, and profiles from the Privacera applications and the cluster, offering valuable insights into system health, performance, and reliability.

The diagram below provides a high-level representation of the monitoring tools installed by Privacera. These tools, sourced from the open-source community, can be customized based on specific requirements. See Advanced Configuration for more details.

Architecture Diagram

** Not all Privacera applications may provide telemetry data.

Note

By default, Loki, Tempo, and Pyroscope are disabled. To enable these components, follow the instructions in Advanced Configuration.

Components & Workflow

  1. Privacera Apps (Python/Java)
    • Some applications running in the system send logs, traces, and profiles to the observability stack via the OTEL Agent.
    • The applications also generate metrics, which are scraped by the OpenTelemetry (OTel) Scraper..
  2. OTel Components
    • OTel Agent (DaemonSet): Runs on every node to collect cluster logs and forward them to Loki.
    • OTel Receiver (Deployment): Receives and processes application standard output logs and traces, forwarding them to Loki, Prometheus, Tempo, and Pyroscope.
    • OTel Scraper (Deployment): Scrapes Privacera application metrics and forwards them to Prometheus.
  3. Infrastructure Metrics Collection
    • Prometheus Node Exporter (DaemonSet): Collects cluster-level system metrics, such as CPU, memory, and disk usage, and forwards them to Prometheus.
  4. Data Storage & Processing
    • Loki: Stores logs for both cluster and application-level logging.
    • Prometheus: Stores metrics from both applications and the cluster.
    • Tempo: Stores tracing data, enabling distributed tracing across microservices.
    • Pyroscope: Captures profiling data for performance analysis.
  5. Visualization & Monitoring
    • Grafana: Connects to Loki, Prometheus, Tempo, and Pyroscope to provide a unified view of logs, metrics, traces, and profiles.

This observability stack helps troubleshoot issues, monitor application health, analyze performance bottlenecks, and ensure system reliability.

Comments