The Monitoring System project is designed to provide continuous surveillance of an operating system by monitoring specific metrics. The system consists of a kernel and dynamically connectable agents, implemented as dynamic libraries. The project is written in C++17 following the Google style guide and employs the Model-View-Controller (MVC) pattern.
- Language: C++17
- Directory Structure:
src: Contains the program code.agents: Dynamic libraries (agents) are stored here.logs: Log files are stored here, each named with the full date of its creation.
- Build System: Makefile with standard GNU targets (all, install, uninstall, clean, dvi, dist, tests).
- Namespace: The program code is organized within the
s21namespace. - Unit Tests: Full coverage of modules with unit tests using the GTest library.
- GUI: Implementation based on any GUI library with C++17 API (e.g., Qt, SFML, GTK+).
- MVC Pattern:
- Business logic is separated into the model.
- No business code in the view.
- No interface code in the controller and model.
- Controllers are thin.
- Dynamic Connection/Disconnection: Agents can be dynamically connected/disconnected while the monitoring system kernel is running.
- Agent Types:
- CPU Load Monitoring Agent
- Memory Monitoring Agent
- Network Monitoring Agent
- Agent Configuration:
- Agent name
- Agent type
- List of critical metric values
- Metrics update time
- Metrics are logged in a specified format with a timestamp.
- Separate log file for each day of system operation.
- Display the last 20 lines of the log with dynamic updates.
- List of connected agents.
- Detailed information about agents:
- Agent type
- Monitored metrics
- Time elapsed since agent's start
- Metrics update time
- Dynamic configuration changes for running agents with saving to a configuration file.
- Disconnecting selected agents.
- Simple Telegram bot for sending messages on critical metric values.
- Enable/disable duplication of notifications to a specified email address.
Implement at least one more agent collecting the following additional metrics:
- CPU load by privilege level
- Total swap volume
- Amount of swap used
- Number of processes ready to run in the queue
- Counting full and free virtual memory
- Total number of inodes
- Average hard disk read time
- Number of errors from the system log
- Number of user authorizations
-
Clone the repository:
git clone https://github.com/mrdamos69/MonitoringSystem.git cd MonitoringSystem -
Build the project:
make install
Dmitrii Chichkanov (mrdamos69)