SystemMetricsPanel

An overview of a lightweight Bash-based system monitor

Abstract

SystemMetricsPanel is a self-contained Bash module for low-overhead observation of host-level performance indicators over Bash shell.
Originally implemented as a single-line “monitoring bar,” the system has evolved into a multi-line framed dashboard, providing a clearer and more expressive representation of system state while maintaining minimal resource consumption.

It exposes two complementary execution modes:

1. Realtime mode, which refreshes an in-place textual panel at a fixed cadence, and
2. Snapshot mode, which emits a single multi-line report suitable for logging or embedding into other tools.

The script relies exclusively on standard GNU/Linux interfaces such as /proc, /sys, and df, ensuring maximum portability and zero external dependencies.

---

1. Design goals

The script is designed around the following objectives:

• Portability – Runs on any common Linux distribution using only POSIX tools.
• Low overhead – Sampling incurs negligible system load.
• SSH compatibility – Designed for terminal-based remote sessions.
• Self-configuration – Automatically detects a suitable network interface.
• Clear separation of logic – Data gathering and presentation layers remain distinct.

---

2. Functional overview

Each refresh cycle performs the following steps:

1. Determine the network interface
2. Sample system metrics
   • CPU load
   • Memory usage
   • Network throughput
   • Disk utilization
   • Host metadata
3. Normalize and compute meaningful statistics
4. Render the panel
   • Multi-line Unicode frame
   • Colors via ANSI sequences
   • CPU mini-bar (text-based)
5. Mode behavior
   • Realtime: refresh in-place
   • Snapshot: print once and exit

---

3. Metric computation details

3.1 CPU utilization (delta method)

Two /proc/stat samples are taken one second apart:

• Compute total and idle times
• CPU% = (1 - Δidle / Δtotal) * 100
• Rendered as:

  CPU  : 23% [####------]
  

3.2 Memory usage

Derived from /proc/meminfo:

• MemTotal - MemAvailable
• Converted to GiB
• Example:

  MEM  : 3.12 GB / 7.80 GB
  

3.3 Network throughput

Reads /sys/class/net/<iface>/statistics/*_bytes:

• Computes deltas over the interval
• Converts bytes → Mbps
• Example:

  NET  : ↑2.44 Mb/s | ↓0.97 Mb/s
  

3.4 Disk utilization

Uses df -hP, extracts % for / and /boot/efi:

DISK : /: 42%, /boot/efi: 12%

---

4. Execution modes

4.1 Snapshot mode

Single printed panel → exit.

4.2 Realtime mode

Updates continuously:

• Recomputes metrics
• Moves cursor up
• Redraws entire panel
• Quit with q

---

5. Usage

chmod +x SystemMetricsPanel.sh

# Default (snapshot)
./SystemMetricsPanel.sh

# Realtime (if script prompts for mode)
./SystemMetricsPanel.sh

# Specify interface
./SystemMetricsPanel.sh eth0

---

6. Representative output

6.1 Realtime example

┌────────────────────────────────────────────────────────────────────────────┐
│                          SYSTEM METRICS PANEL (LIVE)                       │
├────────────────────────────────────────────────────────────────────────────┤
│HOST : srv-node01                                                           │
│USER : admin                                                                │
│UP   : 152 h                                                                │
│CPU  : 17% [###-------]                                                     │
│MEM  : 2.91 GB / 7.80 GB                                                    │
│NET  : ↑3.12 Mb/s | ↓1.02 Mb/s                                              │
│DISK : /: 42%, /boot/efi: 12%                                               │
│Press q to quit                                                             │
└────────────────────────────────────────────────────────────────────────────┘

6.2 Snapshot example

┌────────────────────────────────────────────────────────────────────────────┐
│                           SYSTEM METRICS SNAPSHOT                          │
├────────────────────────────────────────────────────────────────────────────┤
│HOST       : srv-node01                                                     │
│USER       : admin                                                          │
│UPTIME     : 152 h                                                          │
├────────────────────────────────────────────────────────────────────────────┤
│CPU LOAD   : 23% [####------]                                               │
│MEMORY     : 3.12 GB / 7.80 GB                                              │
│NETWORK    : ↑2.44 Mb/s | ↓0.97 Mb/s                                        │
│DISK       : /: 42%, /boot/efi: 12%                                         │
└────────────────────────────────────────────────────────────────────────────┘

---

7. Extensibility

Possible future enhancements:

• Per-core CPU view
• Historical plot storage
• Temperature sensors
• JSON/CSV export

---

8. Licensing and disclaimer

Provided “as-is,” without warranty.
You may freely adapt or extend the script.