Configure metrics and Grafana dashboards

[NGirchev]

Goal

Set up useful application metrics and Grafana dashboards for OpenDaimon runtime observability.

Scope

• Review existing Micrometer/OpenDaimon metrics and identify gaps.
• Expose/verify Prometheus-compatible metrics endpoint for the app runtime.
• Configure Grafana dashboard(s) for core operational signals.
• Add panels for at least:
  • request/message processing rate;
  • processing latency;
  • AI request latency and failures;
  • Telegram/REST request volume if available;
  • bulkhead/priority queue health if available;
  • JVM memory/GC/thread basics;
  • database connection pool health.
• Add basic alert recommendations or dashboard thresholds for failures/latency/resource pressure.
• Document how to run Grafana/Prometheus locally and how to import/use dashboards.

Acceptance Criteria

• Local Docker setup can start metrics stack together with the app or with documented commands.
• Grafana shows live metrics from the running app.
• Dashboard JSON or provisioning files are committed to the repository.
• Documentation explains the metrics endpoint, Grafana URL, and how to validate data flow.
• Verification commands are documented or automated.

Notes

Prefer existing project style and current observability components. Keep the first iteration pragmatic: enough dashboards to diagnose production issues, without overbuilding a full monitoring platform.