Overview

Probestyx is a lightweight metrics collection agent designed to scrape, aggregate, and expose system and custom metrics through a simple HTTP API. It's a standalone metrics-gathering agent but still works seamlessly with the Logstyx logging platform.

What is Probestyx?

Probestyx is a standalone service that runs on your servers to collect metrics from multiple sources:

• System metrics: CPU, memory, disk, network, and process statistics
• Custom metrics: Pull data from URLs, local files, or other applications
• Third-party metrics: Scrape Prometheus exporters, JSON APIs, or text-based outputs

All collected metrics are aggregated and exposed at a single HTTP endpoint (/metrics), making it easy to poll from monitoring systems like Logstyx.

Key Features

🔧 Flexible Metric Collection

• Built-in system metrics for CPU, RAM, disk, network, and processes
• Multiple source types: HTTP endpoints, local files, and shell commands (e.g. docker logs)
• Multiple format support: JSON, Prometheus exposition format, raw text parsing, and log-line grepping
• Metric transformations: Apply calculations to modify values (e.g., convert bytes to MB)
• Error log monitoring: Tail an nginx error log or docker logs output for lines matching a pattern, reporting only what's new since the last scrape — see Scrapers → Log Parser

🎯 Easy Integration

• Single endpoint: All metrics available at /metrics
• JSON output: Easy to parse and integrate
• Optional authentication: HMAC-based signing for secure communication
• Logstyx-ready: Works out-of-the-box with Logstyx probes

⚡ Lightweight & Fast

• Small footprint: Minimal resource usage
• Written in Go: Fast, compiled binary with no runtime dependencies
• Configurable caching: Reduce load on scraped sources
• Pattern filtering: Include/exclude metrics to reduce payload size

🚀 Production Ready

• Cross-platform: Linux, macOS, Windows support
• Systemd/Launchd/NSSM: Runs as a system service with auto-restart
• Docker support: Container-ready with official images
• Easy installation: One-command install scripts for all platforms

How It Works

┌─────────────────────────────────────────────────────────┐
│                      Probestyx Agent                     │
│                                                          │
│  ┌────────────────┐                                     │
│  │ System Metrics │──┐                                  │
│  │  • CPU         │  │                                  │
│  │  • RAM         │  │                                  │
│  │  • Disk        │  │         ┌──────────────┐        │
│  │  • Network     │  ├────────►│  Aggregator  │        │
│  └────────────────┘  │         │              │        │
│                      │         │  • Collect   │        │
│  ┌────────────────┐  │         │  • Transform │        │
│  │ Custom Sources │  │         │  • Filter    │        │
│  │  • JSON APIs   │──┤         └──────┬───────┘        │
│  │  • Prom Files  │  │                │                │
│  │  • Text Files  │  │                │                │
│  └────────────────┘  │                ▼                │
│                      │         ┌──────────────┐        │
│  ┌────────────────┐  │         │ HTTP Server  │        │
│  │ External Apps  │  │         │ :9100        │        │
│  │  • Node Export │──┘         │              │        │
│  │  • App APIs    │            │ GET /metrics │◄───────┼──── Logstyx
│  └────────────────┘            │ GET /health  │        │     Probe
│                                └──────────────┘        │
└─────────────────────────────────────────────────────────┘

Use Cases

Server Monitoring

Deploy Probestyx on each server to collect system metrics and send them to Logstyx for centralized monitoring and alerting.

system:
  enabled: true
  metrics:
    - cpu_usage_percent
    - ram_usage_percent
    - disk_usage_percent
    - network_bytes_sent_per_sec

Application Metrics

Collect custom metrics from your applications by pointing Probestyx at your app's metrics endpoint.

scrapers:
  - name: app_metrics
    source:
      type: url
      url: "http://localhost:8080/stats"
      format: json
    metrics:
      - path: "requests.total"
        name: "total_requests"

Multi-Source Aggregation

Combine metrics from multiple sources (system, Prometheus exporters, custom apps) into a single endpoint.

system:
  enabled: true
  metrics:
    - cpu_usage_percent

scrapers:
  - name: node_exporter
    source:
      type: url
      url: "http://localhost:9100/metrics"
      format: prometheus
    metrics:
      - match: "node_memory_.*"
        
  - name: app_stats
    source:
      type: file
      path: "/var/app/stats.json"
      format: json

Architecture

Probestyx follows a simple, modular architecture:

• System Collector: Native Go libraries to collect OS metrics
• Scrapers: Pluggable collectors for external sources
• Parser: Supports JSON, Prometheus, and regex-based parsing
• Transformer: Apply calculations and filters to metrics
• HTTP Server: Exposes aggregated metrics via REST API
• Optional Auth: HMAC-SHA256 request signing

Comparison with Alternatives

Feature	Probestyx	Prometheus Node Exporter	Telegraf
Primary Purpose	Logstyx integration	Prometheus metrics	InfluxDB/general
Output Format	JSON	Prometheus	Multiple
Configuration	YAML	Flags	TOML
Custom Metrics	✅ Built-in scraping	❌ Separate exporters	✅ Input plugins
Metric Transformation	✅ Calculate field	❌ No	✅ Processors
Authentication	✅ Optional HMAC	❌ No	✅ Various
Installation	Single binary + script	Single binary	Single binary
Binary Size	~10-15 MB	~20 MB	~50-100 MB

Choose Probestyx if:

• You're using Logstyx for centralized logging
• You want JSON output with minimal configuration
• You need to aggregate metrics from multiple sources
• You want built-in metric transformations

Choose alternatives if:

• You're deeply invested in Prometheus ecosystem
• You need advanced metric processing pipelines
• You require native InfluxDB line protocol

Performance

Probestyx is designed to be lightweight and efficient:

• Memory footprint: 10-30 MB typical usage
• CPU usage: <1% idle, 2-5% during collection
• Collection speed: <100ms for typical configurations
• Concurrent scrapers: Parallel execution for faster aggregation

Security

• Optional authentication: Enable HMAC signing to prevent unauthorized access
• No data storage: Metrics are collected and served, not stored
• Read-only operations: No write capabilities to monitored systems
• Configurable timeouts: Prevent hanging on slow sources

Support & Community

• Source Code: GitHub Repository
• Issues: Report a bug

Next Steps

• Installation - Install Probestyx on your system
• Configuration - Learn how to configure Probestyx
• System Metrics - Available system metrics reference
• Integration - Connect Probestyx to Logstyx