A terminal-based system monitor for ROS2 environments with real-time system metrics, ROS node/topic monitoring, and TF frame tracking.

• Version 0.1 - Still has bugs and design flaws
• Verified to work running on Ubuntu 24.04
• Requires separate insteallation of python library textual

• ROS2 (Humble, Iron, or Jazzy)
• Python 3.8+
• Python library textual

You may find that pip install textual gives an error error: externally-managed-environment. I have found that if you can get around this with pip install textual --break-system-packages. NB: I am not positive how that works and can't guarantee that it is safe. It's worked for me and several otehrs.

Like many other ROS2 apps:

cd ros2_ws/src
git clone https://github.com/pitosalas/ros2sysmon.git

cd ~/your_ros2_workspace
source ~/your_ros2_workspace/setup.basj
colcon build --packages-select ros2sysmon
source install/setup.bash

# Basic usage
ros2 run ros2sysmon ros2sysmon

# With options
ros2 run ros2sysmon ros2sysmon --config /path/to/config.yaml --refresh-rate 3.0

• 1 - Topics + TF frames view
• 2 - ROS nodes + processes view
• r - Manual refresh/collect data
• x/q - Exit

System metrics with ASCII progress bars:

• CPU/Memory/Disk usage with visual bars
• Load average, uptime, temperature
• Network latency to 8.8.8.8

1. Mode 1: ROS Topics (Hz, Count) + TF Frames
2. Mode 2: ROS Nodes + System Processes (filtered for ROS)

System alerts with timestamps for threshold violations.

• CPU: psutil.cpu_percent(interval=0.1) - 100ms sampling
• Memory: psutil.virtual_memory().percent - available vs total
• Disk: psutil.disk_usage('/').percent - root filesystem usage
• Temperature: psutil.sensors_temperatures() - from coretemp/cpu_thermal
• Load Average: psutil.getloadavg()[0] - 1-minute load
• Uptime: time.time() - psutil.boot_time() formatted as HH:MM
• Network Latency: ping -c 1 -W 2 8.8.8.8 parsed from output

• Nodes: Discovered via ros2 node list (10s timeout)
• Topics: Discovered via ros2 topic list -t (5s timeout)
• Topic Hz:
  • Creates ROS2 subscribers for all available topics
  • Collects message timestamps during windowed collection (default 3s)
  • Calculates frequency as: message_count / collection_window_duration
  • Shows 2 decimal precision
• TF Frames: Retrieved via tf2_ros.Buffer.all_frames_as_yaml()
• Processes: Filters system processes for ROS-related keywords

• /cmd_vel, /twist: 10Hz
• /odom: 30Hz
• /scan, lidar topics: 10Hz
• /image*, camera topics: 30Hz
• /imu: 100Hz
• /tf: 100Hz

Default config: share/ros2sysmon/config/default_config.yaml

• system_metrics: 5.0s - CPU, memory, disk, temperature
• network_ping: 5.0s - Network latency checks
• ros_discovery: 10.0s - Node/topic discovery
• hz_collection_duration: 3.0s - Hz measurement window

• CPU: Warn 70%, Error 85%
• Memory: Warn 75%, Error 90%
• Disk: Warn 80%, Error 95%
• Temperature: Warn 70°C, Error 85°C
• Network: Warn 100ms, Error 500ms

• Textual framework for terminal UI
• Threading: Separate collectors for system, network, and ROS data
• SharedDataStore: Thread-safe data exchange
• Manual/Timed Collection: Configurable collection intervals

ROS2: rclpy, tf2_ros, rosidl_runtime_py
System: psutil, pyyaml, textual
Standard: subprocess, threading, dataclasses

MIT