🖥️ System Health Analyzer (Linux)

A small CLI-based system monitoring tool that captures raw OS signals and turns them into clear health decisions.

This project focuses on resource pressure, not dashboards or raw metrics.

---

📖 Overview

The system is built as a simple pipeline:

Collect
   │
   ▼
Snapshot
   │
   ▼
Analyze

Workflow

1. Collectors (Bash) read low-level system data.
2. Coordinator (Python) merges data into a single snapshot.
3. Analyzer (Python) evaluates system health and emits decisions.

The result is a monitoring tool suitable for automation, not just observation.

---

📂 Repository Structure

system-health-analyzer/
├── analyzer/
│   └── analyze.py
├── collectors/
│   ├── disk_collector.sh
│   ├── memory_collector.sh
│   └── process_collector.sh
├── data/
│   └── metrics.json
└── coordinator.py

---

⚙️ Components

1️⃣ Collectors (Bash)

Located in collectors/

Each collector:

• Reads system state directly from the OS
• Emits valid JSON to stdout
• Contains no thresholds or decisions

Included Collectors

Collector	Description
memory_collector.sh	Reads /proc/meminfo
disk_collector.sh	Reads df
process_collector.sh	Reads ps

---

2️⃣ Coordinator (Python)

The coordinator:

• Runs all collectors
• Parses their JSON output
• Merges data into one snapshot
• Writes the snapshot to data/metrics.json

This ensures data integrity and prevents malformed JSON.

---

3️⃣ Analyzer (Python)

The analyzer:

• Loads the snapshot
• Applies health rules
• Emits OK / WARNING states
• Returns automation-safe exit codes

Exit Codes

Code	Meaning
0	Healthy
1	Warning (resource pressure)
2	Invalid or incomplete data

---

🚀 How to Use

1. Generate a system snapshot

python3 coordinator.py

2. Analyze system health

python3 analyzer/analyze.py

---

🎯 Design Principles

• Clear separation between data collection and decision logic
• Raw OS signals over synthetic metrics
• JSON as a strict data contract
• Fail fast on invalid data
• Designed for cron, CI, or automation hooks

---

💡 Why This Project?

This project was built to understand:

• How Linux exposes real system pressure
• Why monitoring should drive decisions, not dashboards
• How to design small, composable system tools reopening

---

📝 Notes

• Developed and tested on Arch Linux
• Intentionally minimal by design
• Built as a systems-learning artifact, not a full monitoring suite