A high-performance, concurrent remote diagnostic agent designed for Linux-based systems. This solution allows system administrators to monitor real-time process telemetry—including memory consumption, CPU load, and network resources—across a network using a custom-engineered binary protocol.
Instead of relying on heavy shell-based tools, this project interfaces directly with the Linux Kernel via the ProcFS (/proc) virtual filesystem using low-level system calls, ensuring a minimal resource footprint on the target system.
Concurrent Server Architecture: Implemented a multi-process model using the fork() API, enabling the server to handle multiple Network Operations Center (NOC) clients simultaneously without blocking.
Direct ProcFS Parsing: Manual extraction of kernel-level metrics using raw system calls (open, read, readlink):
VmRSS Tracking: Monitors actual physical memory (Resident Set Size) from /proc/[pid]/status.
Differential CPU Sampling: Calculates CPU load delta using utime and stime ticks over a sample window from /proc/[pid]/stat.
FD & Socket Resolution: Counts open file descriptors and identifies active network sockets via symbolic link resolution in /proc/[pid]/fd.
Persistent Session History: Manages a sliding-window command history (last 50 commands) with state persistence to a hidden .remote_history file.
Remote Process Control: Integrated process termination capability using the SIGKILL signal via the kill() system call.
Robust Buffer Safety: Utilizes boundary-checked string formatting (snprintf) across all diagnostic functions to mitigate buffer overflow vulnerabilities.
This project follows a clean, professional, and scalable directory structure:
.
├── bin/ # Compiled executables (Server/Client)
├── include/ # Header files (API Definitions & Macros)
│ ├── common.h # Shared constants (MAX_BUF, MAX_HISTORY)
│ ├── history.h # History module interface
│ ├── network_utils.h # Socket abstraction layer
│ └── proc_utils.h # Linux Internals / ProcFS interface
├── src/ # Implementation files
│ ├── client_main.c # Interactive diagnostic terminal
│ ├── history.c # Circular-buffer history logic
│ ├── network_utils.c # TCP connection & protocol handling
│ ├── proc_utils.c # Low-level kernel data extraction
│ └── server_main.c # Multi-process agent core
├── .remote_history # Hidden persistence file
├── Makefile # Automated build system
└── test_suite.sh # Automated integration & regression test script
| Category | Tool/Standard |
|---|---|
| Language | C (C11) |
| IPC | POSIX Sockets (TCP/IP) |
| Concurrency | Multi-processing (fork) |
| VFS Access | Linux ProcFS (/proc) |
| Build System | GNU Make |
| Testing | Custom Shell Integration Suite |
| Version Control | Git (Semantic Tagging v1.0) |
This is the easiest and most reliable way to build.
-
Build the project:
The project uses a centralized Makefile for streamlined compilation:
make clean && make -
Launch the Agent (Server)
Start the diagnostic server on the target machine (specify a port, e.g., 8080):
./bin/server 8080
-
Connect the Console (Client)
On the monitoring machine, connect to the agent's IP and port:
./bin/client 127.0.0.1 8080
Planned enhancements to increase security and system depth:
Transport Layer Security: Implement OpenSSL/TLS to encrypt the command channel.
Containerization: Add Multi-stage Dockerfiles for zero-install deployment.
Advanced Metrics: Support for /proc/net/dev to monitor real-time network bandwidth.
User Interface: Implement Up/Down arrow key history navigation using termios raw mode.
Graceful Shutdown: Implement advanced signal handling for the server to clean up child processes on SIGINT.
Contributions are what make the open-source community such an amazing place to learn, inspire, and create. Any contributions you make are greatly appreciated.
- Fork the Project
- Create your Feature Branch (
git checkout -b feature/AmazingFeature) - Commit your Changes (
git commit -m 'feat: Add some AmazingFeature') - Push to the Branch (
git push origin feature/AmazingFeature) - Open a Pull Request
Distributed under the MIT License. See LICENSE for more information.