obsv

Simple Observability Platform

Deployment

All services are packaged as an all-in-one executable, which can be configured to serve only specific services (deployed separately).

cargo run --bin osv

Overview

flowchart LR
    in_http["HTTP (Otlp, events)"]
    in_grpc[Otel GRPC]
    in_services[Services]

    out_api[API]
    out_dash[Dashboard]
    out_status[Status page]
    out_notifs[Notifications]

    subgraph obsv [Obsv]
        coll[Collector]
        monitor[Monitor]
        db["DB (ClickHouse)"]
        api[Data API]
        dash[Dashboard]
        status[Status page]
        notify[Notifier]
    end

    in_grpc --> |:4317| coll
    in_http --> |:4318| coll
    in_services --> monitor

    coll --> db
    db --> api
    monitor --> db
    db --> notify

    api --> |:5001| out_api
    dash --> |:5002| out_dash
    status --> |:5003| out_status
    notify --> |emails, hooks| out_notifs

Loading

Repo organization

• obsv-core: core structures and utilities
• obsv-collect: collector service
• obsv-monitor: monitoring service
• obsv-api: data API service
• obsv-notify: notification service
• obsv-dash: dashboard service
• obsv-status: status page service
• obsv-bin: executable
• lib/**: shared libraries

Similar tools

• Jaeger: tracing collector and dashboard
• Prometheus: metrics and alerting toolkit
• teletrace: tracing collector and dashboard (ElasticSearch DB + Collector/Server).
• Vigil: monitors, alerts, status page
• QRYN: observability platform
• ...