Phobos Dashboards

Overview

Stop flying blind in production. These Grafana dashboards give you instant visibility into your Akka.NET cluster's actual behavior - see which actors are struggling, where messages are piling up, and what's causing those 3am alerts.

🎯 Why Prometheus & Grafana?
Same stack Microsoft uses for ASP.NET Core. Same stack Netflix uses. Same stack your DevOps team already knows. We didn't reinvent the wheel - we built on the industry standard that's monitoring millions of production systems right now. Your metrics work everywhere because they're OpenTelemetry, but these dashboards are the fastest path to production insight.

What You Get

  • Find bottlenecks before your users do - See exactly which actors are slow, which mailboxes are backing up, and where your cluster is struggling
  • Debug production issues in seconds - Real actor counts, message rates, and error breakdowns mean you can diagnose problems without diving into logs
  • Import and go - Two clicks in Grafana Cloud. That's it. No custom queries, no dashboard building, no guesswork
  • Works with your existing stack - OpenTelemetry metrics export to DataDog, New Relic, Azure Monitor, or anywhere else you're already paying for

Available Dashboards

Source Code: https://github.com/petabridge/phobos-dashboards

Grafana Cloud Dashboards:

🌐 Cluster Overview Dashboard

Monitor your entire Akka.NET cluster's health and performance metrics at a glance.

Cluster Status and Node Health

Cluster Status & Node Health

Real-time cluster membership tracking with node status, message rates, and instance details

Logging and Error Tracking

Logging & Error Tracking

Monitor logging activity, exception rates, and error type breakdowns across the cluster

Actor Metrics

Actor Metrics & Backpressure

Track actor message processing, mailbox backlogs, and throughput per node

Sharding Statistics

Sharding Statistics

Detailed sharding distribution with message type breakdowns and throughput analysis

⚡ Actor Performance Dashboard

Deep-dive into individual actor performance metrics and message processing patterns.

**Grafana Cloud**: **GitHub Source**:
Actor Overview and Key Metrics

Actor Overview & Key Metrics

Real-time actor counts, message volumes, throughput rates, and latency heatmaps

Detailed Performance Analysis

Detailed Performance Analysis

Message processing latencies, mailbox depths, and per-instance throughput metrics

Installation

Prerequisites

Quick Import from Grafana Cloud

  1. Navigate to Dashboards → Import in your Grafana instance
  2. Enter the dashboard ID:
    • 15637 for Cluster Overview Dashboard
    • 15638 for Actor Performance Dashboard
  3. Select your Prometheus data source
  4. Click Import

Manual Installation

  1. Download the dashboard JSON files from: https://github.com/petabridge/phobos-dashboards
  2. Navigate to Dashboards → Import in Grafana
  3. Upload or paste the JSON content
  4. Select your Prometheus data source
  5. Click Import

Key Metrics Reference

Metric Description Dashboard
Actor System Metrics
akka_actor_live_actors Number of active actors in the system Both
akka_actor_restarts_total Total actor restarts (indicates failures) Both
akka_actor_mailbox_depth_msgs Messages waiting in actor mailboxes Both
Message Flow Metrics
akka_messages_recv_msgs_total Total messages received by actors Both
akka_messages_latency_milliseconds_bucket Message processing latency histogram Performance
akka_messages_deadletters_msgs_total Messages sent to dead letters Both
akka_messages_unhandled_msgs_total Unhandled messages Both
Cluster Metrics
akka_cluster_members Current cluster member count Cluster
akka_cluster_reachable_members Reachable cluster members Cluster
akka_address Node address information Cluster
Sharding Metrics
akka_cluster_sharding_entities Number of sharded entities Cluster
akka_cluster_sharding_shards Number of active shards Cluster
Logging Metrics
akka_logs_events_total Log events by severity level Cluster

For a complete list of available metrics, see What Data Does Phobos Capture?

Adapting to Other Platforms

📝 Reference Architecture
These Grafana dashboards serve as a reference implementation. The PromQL queries, panel configurations, and metric selections can be adapted to other monitoring platforms like DataDog, New Relic, or Azure Monitor. Since Phobos exports standard OpenTelemetry metrics, your data is portable across any OpenTelemetry-compatible backend.

Translation Guide

  • Metrics: All akka_* prefixed metrics are available via OpenTelemetry
  • Queries: PromQL queries can be translated to other query languages
  • Visualizations: Panel types and configurations provide blueprint for other platforms
  • Alerts: Threshold values and conditions are platform-agnostic

Troubleshooting

Dashboard shows "No Data"

  1. Verify Prometheus is receiving metrics: curl http://prometheus:9090/api/v1/label/__name__/values | grep akka
  2. Check data source configuration in Grafana
  3. Ensure time range selection includes recent data
  4. Validate Phobos is configured correctly

Missing Panels or Metrics

  1. Confirm Phobos monitoring is enabled
  2. Check OpenTelemetry exporter configuration
  3. Verify Prometheus scrape configuration includes your application
  4. Review application logs for metric export errors

Performance Impact

See our Performance Impact documentation for guidance on optimizing Phobos overhead.

Support & Resources