// PHOBOS

OpenTelemetry for Akka.NET

Phobos is OpenTelemetry integration purpose-built for Akka.NET. Automatic instrumentation, intelligent noise reduction, and built-in dashboards — so you see what matters, not everything.

Buy Now Explore Phobos View Dashboards

// trusted_by

Used by enterprises worldwide

~2M Total Downloads

Enterprise Fortune 500 Companies

// WHY_PHOBOS

OpenTelemetry Native

First-class OpenTelemetry integration. Traces, metrics, and logs flow through standard OTel pipelines to Seq, Grafana, Datadog, Jaeger, or any OTel-compatible backend.

Intelligent Noise Reduction

Akka.NET is extremely noisy — inter-actor messaging is cheap, so tracing produces gigantic data volumes. Phobos automatically curtails noise and surfaces only useful signals. We put a lot of care into tuning accuracy, conciseness, and utility.

Built-in Dashboards

Pre-built dashboards for cluster performance, actor backlogs, throughput, error rates, and entity distribution. Ready to go out of the box. View screenshots →

// SEE_IT_IN_ACTION

Built-in Grafana Dashboards

Phobos ships with pre-built dashboards for Akka.NET Cluster monitoring, including message throughput by type, shard distribution, and entity distribution across nodes.

Phobos Akka.NET Cluster dashboard in Grafana

Phobos Cluster + Sharding dashboard with Prometheus data source

Phobos cluster overview dashboard showing node count and status

Cluster overview — node count, status distribution, and per-instance sparklines

Phobos actor metrics dashboard showing message throughput and mailbox backlog

Actor metrics — message throughput, mailbox backlog, and total volume

Phobos actor performance dashboard showing latency percentiles

Actor performance — P50/P95/P99 latency and message type breakdown

// CUSTOMER_STORY

At Tranzact we use Akka.NET in a cluster made of almost 150 nodes on production... we took the chance to replace our custom trace implementation with Phobos (Jaeger+AI), and it has proved to be a useful tool for troubleshooting and monitoring. It helped us quickly discover race conditions in our software and improve our error handling.