SnapSys

SnapSys is a lightweight, purpose-built CLI tool for capturing CPU, memory, and disk usage at fixed intervals, outputting the results in clean, structured JSONL format. Designed for system administrators, developers, DevOps engineers, and performance testers, SnapSys is ideal for monitoring system behavior, debugging performance issues, or capturing lightweight metrics during CI/CD pipelines or local build processes. Whether you're analyzing resource usage in virtual machines, testing workloads in Docker containers, or gathering evidence for incident reports, SnapSys fits seamlessly into your workflow.

⚠️ SnapSys is currently developed and tested for Linux systems only.

Key Features

• Track CPU, Memory, and Disk usage at defined intervals
• Output in JSONL format for easy processing and analysis
• Minimal dependencies – just Go and your shell
• Includes automatic fallback output path if none is provided

Download a Precompiled Binary

Visit the Releases Page

Download the binary called snapsys

In your terminal run

chmod +x snapsys

Run it:

./snapsys snapshot

To simplify usage:

sudo mv snapsys /usr/local/bin/snapsys
snapsys snapshot

---

Usage

Basic Command

snapsys snapshot

This will collect system stats every 3 seconds for 30 seconds, and write to a default path like:

./snapsys_runs/snaprun_2025-06-16_13-00-00.jsonl

Command Flags

Flag	Default	Description
--duration	30s	Total time to run the snapshot
--interval	3s	Time between each snapshot (minimum 1s)
--output	(auto-generated)	Output file path (.jsonl required if specified)

⚠️ If the interval is set below 1s, SnapSys will default it to 1s and warn the user.

⚠️ If you use --output, the file must end in .jsonl.

---

Use Cases

SnapSys is ideal for:

• Lightweight benchmarking in Linux VMs or containers
• Capturing performance stats during builds or deployments
• Logging system behavior for incident reports
• Analyzing trends across time during load tests

---

Example Output (JSONL)

Each snapshot is written as a line in JSON:

{"timestamp":"2025-06-16T13:00:00Z","cpu":{"usage_pct":12.8, ...},"memory":{"usage_pct":31.9, ...},"disks":{"/":{...}}}

This is great for:

• Ingesting into tools like Elasticsearch
• Plotting in Python, Grafana, etc.
• Please see example.jsonl file for full example

---

Future Plans

• Additional system metrics:

  • Uptime
  • Hostname and kernel version
  • Network interface stats (bytes in/out, errors, drops)
  • System load average

• Snapshot tagging and summaries

• Ability to push .jsonl data to remote HTTP APIs

• Support for compressed output (.jsonl.gz)

• Interactive shell mode for quick inspection and dev workflows

• CSV export format option

---

PRs welcome!