SysMon — Linux System Monitor

SysMon is a real-time console application for monitoring key Linux system metrics including CPU, memory, disk I/O, and network activity. Built with performance and clarity in mind, it leverages native /proc filesystem interfaces and runs efficiently using a thread pool.

Features

• Real-time monitoring of:
  • CPU usage (overall and per-core)
  • Memory usage (including swap)
  • Disk I/O (read/write throughput, IOPS, utilization)
  • Network activity (per-interface throughput)
• Configurable update interval (e.g., 500ms, 2s)
• File-based logging with periodic system summaries
• Multi-threaded architecture using a custom thread pool
• Zero external dependencies — only standard C++17 and Linux system interfaces

Build

Requirements

• A C++17-compliant compiler (e.g., g++ ≥ 7)
• GNU Make
• Linux system with /proc filesystem

Compile

make

This produces the executable sysmon in the project root.

Usage

./sysmon [OPTIONS]

Examples

# Update every 1 second, log to default log.txt
./sysmon

# Update every 300ms, log to custom file
./sysmon -i=300ms -l=monitor.log

# Enable per-core CPU stats
./sysmon --per-core

# Show help or version
./sysmon help
./sysmon version

Command-Line Options

Option	Description
-i=<dur> / --interval=<dur>	Update interval (e.g., 1s, 200ms)
-l=<file> / --log-file=<file>	Log output file (default: log.txt)
--log-interval=<dur>	Interval between full log summaries (default: 120s)
--per-core	Show CPU usage per core
help	Display help message
version	Show version info

Supported duration formats: <N>s (seconds) or <N>ms (milliseconds).

Metrics Collected

• CPU: Total usage %; optionally per-core breakdown.
• Memory: Used vs. total (GiB), % usage, and swap utilization (if enabled).
• Disk: Read/write speed (MiB/s), IOPS, and disk utilization %.
• Network: Receive/transmit speed (MiB/s) per interface (excluding lo).

All disk and network values are averaged over the collection interval.

Logging

• Logs are written to log.txt by default (configurable).
• Includes:
  • Startup messages and errors
  • Full system summary every 2 minutes (adjustable via --log-interval)
• Log entries are timestamped with millisecond precision.

Architecture

• Each metric type is handled by a dedicated collector class (CpuCollector, MemoryCollector, etc.)
• All collectors implement the IMetricCollector interface.
• Metrics are gathered in parallel using a thread pool (ThreadPool.hpp).
• No third-party libraries — pure C++17 and Linux /proc interfaces.

License

This project was created as a technical assignment. Free to use, modify, and distribute.