systant/CLAUDE.md

CLAUDE.md

This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.

Common Commands

Development

# Install dependencies
mix deps.get

# Compile the project
mix compile

# Run in development (non-halt mode)
mix run --no-halt

# Run tests
mix test

# Run specific test
mix test test/systant_test.exs

# Enter development shell (via Nix)
nix develop

# Run both server and dashboard together (recommended)
just dev
# or directly: hivemind

# Run components individually
just server      # or: cd server && mix run --no-halt
just dashboard   # or: cd dashboard && mix phx.server

# Other just commands
just deps        # Install dependencies for both projects
just compile     # Compile both projects
just test        # Run tests for both projects
just clean       # Clean both projects

Production

# Build production release
MIX_ENV=prod mix release

# Run production release
_build/prod/rel/systant/bin/systant start

Architecture Overview

This is an Elixir OTP application that serves as a systemd daemon for MQTT-based system monitoring, designed for deployment across multiple NixOS hosts to integrate with Home Assistant.

Core Components

• Systant.Application (lib/systant/application.ex): OTP application supervisor that starts the MQTT client
• Systant.MqttClient (lib/systant/mqtt_client.ex): GenServer handling MQTT connection, metrics publishing, and command subscriptions
• Systant.MqttHandler (lib/systant/mqtt_handler.ex): Custom Tortoise handler for processing command messages with security validation
• Systant.CommandExecutor (lib/systant/command_executor.ex): Secure command execution engine with whitelist validation and audit logging
• Systant.SystemMetrics (lib/systant/system_metrics.ex): Comprehensive Linux system metrics collection with configuration support
• Systant.Config (lib/systant/config.ex): TOML-based configuration loader with environment variable overrides
• Dashboard.Application (dashboard/lib/dashboard/application.ex): Phoenix LiveView dashboard application
• Dashboard.MqttSubscriber (dashboard/lib/dashboard/mqtt_subscriber.ex): Real-time MQTT subscriber that feeds data to the LiveView dashboard

Key Libraries

• Tortoise: MQTT client library for pub/sub functionality
• Jason: JSON encoding/decoding for message payloads
• Toml: TOML configuration file parsing
• Phoenix LiveView: Real-time dashboard framework

MQTT Behavior

• Publishes comprehensive system metrics (CPU, memory, disk, GPU, network, temperature, processes) to stats topic
• Subscribes to commands topic for incoming events that can trigger user-customizable actions
• Uses hostname-based randomized client ID to avoid conflicts across multiple hosts
• Configurable startup delay (default 5 seconds) before first metrics publish
• Real-time metrics collection with configurable intervals

Configuration System

Systant uses a TOML-based configuration system with environment variable overrides:

• Config File: systant.toml (current dir, ~/.config/systant/, or /etc/systant/)
• Module Control: Enable/disable metric collection modules (cpu, memory, disk, gpu, network, temperature, processes, system)
• Filtering Options: Configurable filtering for disks, network interfaces, processes
• Environment Overrides: MQTT_HOST, MQTT_PORT, SYSTANT_INTERVAL, SYSTANT_LOG_LEVEL

Key Configuration Sections

• [general]: Collection intervals, enabled modules
• [mqtt]: Broker settings, client ID prefix, credentials
• [commands]: Command execution settings, security options
• [[commands.available]]: User-defined command definitions with security parameters
• [disk]: Mount filtering, filesystem exclusions
• [gpu]: NVIDIA/AMD GPU limits and settings
• [network]: Interface filtering, traffic thresholds
• [processes]: Top process limits, sorting options
• [temperature]: CPU/sensor temperature monitoring

Default Configuration

• MQTT Host: mqtt.home (configurable via MQTT_HOST)
• Stats Topic: systant/${hostname}/stats (per-host topics)
• Command Topic: systant/${hostname}/commands (per-host topics)
• Response Topic: systant/${hostname}/responses (command responses)
• Publish Interval: 30 seconds (configurable via SYSTANT_INTERVAL)
• Command System: Enabled by default with example commands (restart, info, df, ps, ping)

NixOS Deployment

This project includes a complete Nix packaging and NixOS module:

• Package: nix/package.nix - Builds the Elixir release using beamPackages.mixRelease
• Module: nix/nixos-module.nix - Provides services.systant configuration options
• Development: Use nix develop for development shell with Elixir/Erlang

The NixOS module supports:

• Configurable MQTT connection settings
• Per-host topic naming using ${config.networking.hostName}
• Environment variable configuration for runtime settings
• Systemd service with security hardening
• Auto-restart and logging to systemd journal

Dashboard

The project includes a Phoenix LiveView dashboard (dashboard/) that provides real-time monitoring of all systant instances.

Dashboard Features

• Real-time host status updates via MQTT subscription
• LiveView interface showing all connected hosts
• Automatic reconnection and error handling

Dashboard MQTT Configuration

• Subscribes to systant/+/stats to receive updates from all hosts
• Uses hostname-based client ID: systant-dashboard-${hostname} to avoid conflicts
• Connects to mqtt.home:1883 (same broker as systant instances)

Important Implementation Notes

• Tortoise Handler: The handle_message/3 callback must return {:ok, state}, not []
• Topic Parsing: Topics may arrive as lists or strings, handle both formats
• Client ID Conflicts: Use unique client IDs to prevent connection instability

Development Roadmap

Phase 1: System Metrics Collection (Completed)

• ✅ SystemMetrics Module: server/lib/systant/system_metrics.ex - Comprehensive metrics collection
• ✅ CPU Metrics: Load averages (1/5/15min) via /proc/loadavg
• ✅ Memory Metrics: System memory data via /proc/meminfo with usage percentages
• ✅ Disk Metrics: Disk usage and capacity via df command with configurable filtering
• ✅ GPU Metrics: NVIDIA (nvidia-smi) and AMD (rocm-smi) GPU monitoring with temperature, utilization, memory
• ✅ Network Metrics: Interface statistics via /proc/net/dev with traffic filtering
• ✅ Temperature Metrics: CPU temperature and lm-sensors data via system files and sensors command
• ✅ Process Metrics: Top processes by CPU/memory via ps command with configurable limits
• ✅ System Info: Uptime via /proc/uptime, kernel version, OS info, Erlang runtime data
• ✅ MQTT Integration: Real metrics published with configurable intervals replacing simple messages
• ✅ Configuration System: Complete TOML-based configuration with environment overrides
• ✅ Dashboard Integration: Phoenix LiveView dashboard with real-time graphical metrics display

Implementation Details

• Uses Linux native system commands and /proc filesystem for accuracy over Erlang os_mon
• Configuration-driven metric collection with per-module enable/disable capabilities
• Advanced filtering: disk mounts/types, network interfaces, process thresholds
• Graceful error handling with fallbacks when commands/files unavailable
• JSON payload structure: {timestamp, hostname, cpu, memory, disk, gpu, network, temperature, processes, system}
• Dashboard displays metrics as progress bars and cards with color-coded status indicators
• TOML configuration with environment variable overrides for deployment flexibility

Phase 2: Command System (Completed)

• ✅ Command Execution: server/lib/systant/command_executor.ex - Secure command processing with whitelist validation
• ✅ MQTT Handler: server/lib/systant/mqtt_handler.ex - Custom Tortoise handler for command message processing
• ✅ User Configuration: Commands fully configurable via systant.toml with security parameters
• ✅ MQTT Integration: Commands via systant/{hostname}/commands, responses via systant/{hostname}/responses
• ✅ Security Features: Whitelist-only execution, parameter validation, timeouts, comprehensive logging
• ✅ Built-in Commands: list command shows all available user-defined commands

Command System Features

• User-Configurable Commands: Define custom commands in systant.toml with triggers, allowed parameters, timeouts
• Enterprise Security: No arbitrary shell execution, strict parameter validation, execution timeouts
• Simple Interface: Send {"command":"trigger","params":[...]}, receive structured JSON responses
• Request Tracking: Auto-generated request IDs for command/response correlation
• Comprehensive Logging: Full audit trail of all command executions with timing and results

Example Command Usage

# Send commands via MQTT
mosquitto_pub -t "systant/hostname/commands" -m '{"command":"list"}'
mosquitto_pub -t "systant/hostname/commands" -m '{"command":"info"}'
mosquitto_pub -t "systant/hostname/commands" -m '{"command":"df","params":["/home"]}'
mosquitto_pub -t "systant/hostname/commands" -m '{"command":"restart","params":["nginx"]}'

# Listen for responses
mosquitto_sub -t "systant/+/responses"

Phase 3: Home Assistant Integration (Completed)

• ✅ MQTT Auto-Discovery: server/lib/systant/ha_discovery.ex - Publishes HA discovery configurations for automatic device registration
• ✅ Device Registration: Creates unified "Systant {hostname}" device in Home Assistant with comprehensive sensor suite
• ✅ Sensor Auto-Discovery: CPU load averages, memory usage, system uptime, temperatures, GPU metrics, disk usage, network throughput
• ✅ Configuration Integration: TOML-based enable/disable with homeassistant.discovery_enabled setting
• ✅ Value Templates: Proper JSON path extraction for nested metrics data with error handling
• ✅ Real-time Updates: Seamless integration with existing MQTT stats publishing - no additional topics needed

Home Assistant Integration Features

• Automatic Discovery: No custom integration required - uses standard MQTT discovery protocol
• Device Grouping: All sensors grouped under single "Systant {hostname}" device for clean organization
• Comprehensive Metrics: CPU, memory, disk, GPU (NVIDIA/AMD), network throughput, temperature, and system sensors
• Configuration Control: Enable/disable discovery via systant.toml configuration
• Template Flexibility: Advanced Jinja2 templates handle optional/missing data gracefully
• Topic Structure: Discovery on homeassistant/#, stats remain on systant/{hostname}/stats

Setup Instructions

1. Configure MQTT Discovery: Set homeassistant.discovery_enabled = true in systant.toml
2. Start Systant: Discovery messages published automatically on startup (1s after MQTT connection)
3. Check Home Assistant: Device and sensors appear automatically in MQTT integration
4. Verify Metrics: All sensors should show current values within 30 seconds

Available Sensors

• CPU: Load averages (1m, 5m, 15m), temperature
• Memory: Usage percentage, used/total in GB
• Disk: Root and home filesystem usage percentages
• GPU: NVIDIA/AMD utilization, temperature, memory usage
• Network: RX/TX throughput in MB/s for primary interface (real-time bandwidth monitoring)
• System: Uptime in hours, kernel version, online status

Future Plans

• Multi-host deployment for comprehensive system monitoring
• Advanced alerting and threshold monitoring
• Historical data retention and trending