🛡️ Validator-AI Monitor

An AI-powered self-monitoring agent designed to predict, analyze, and mitigate risks that could cause missed blocks for validator nodes.

---

📋 What This Bot Does

• Collects live system metrics (CPU load, memory, disk IO, network errors) using Prometheus
• Predicts probability of system failure using a trained machine learning model (Gradient Boosted Trees)
• Detects real anomalies (high CPU, low memory, high IO, etc.)
• Automatically mitigates issues safely when possible (e.g., clearing memory cache)
• Sends real-time Telegram alerts if a critical threshold is crossed
• Retrains itself over time based on collected real-world data
• Runs as a lightweight daemon without interfering with your validator performance

---

🚀 How to Run It

1. Clone the repository

git clone https://github.com/GlibxNodes/GlibxNodesAIBot.git
cd GlibxNodesAIBot

2. Install Python Dependencies

pip install -r requirements.txt

requirements.txt contains:
prometheus-api-client, scikit-learn, pandas, python-dotenv, requests, joblib

3. Create a .env file

Create a .env file in the root directory:

PROMETHEUS_URL=http://localhost:9090
TELEGRAM_BOT_TOKEN=your-telegram-bot-token
TELEGRAM_CHAT_ID=your-chat-id

📌 Notes:

• Get your Telegram Bot Token from @BotFather
• Find your Chat ID from @userinfobot

---

4. Run the Bot

To run manually:

python3 validator-ai.py

Or run as a systemd service:

sudo systemctl start validator-ai
sudo systemctl enable validator-ai

---

⚙️ Monitored Metrics

Metric	Purpose
node_load1	CPU system load
node_memory_MemAvailable_bytes	Free memory
node_disk_io_time_seconds_total	Disk I/O pressure
node_network_receive_errs_total	Network errors

---

🔥 Alert Conditions

• CPU load > 5.0 (sustained)
• Available memory < 1GB
• Disk I/O overload
• Rising network errors
• AI-detected anomalies

---

🛡️ Automatic Mitigations

• Memory cache clearing (sync; echo 3 > /proc/sys/vm/drop_caches)
• Process priority adjustment
• Zombie process cleanup (future planned)

---

💬 Telegram Alerts Include

• ✅ Current system metrics snapshot
• ⚠️ Detected primary issue
• 🛠️ Suggested recommended actions
• 🔧 Mitigations applied (if any)

---

❗ Important Notes

• Only safe, read-only system interventions (no validator force restarts)
• Minimal resource usage (~30MB RAM)
• Self-learning model that improves over time
• Always fallback to safe prediction mode if needed

---

📈 Roadmap

flowchart TD
    A[Start Monitoring] --> B{Risk Detected?}
    B -- No --> A
    B -- Yes --> C{Real Anomaly?}
    C -- No --> A
    C -- Yes --> D[Send Telegram Alert]
    D --> E{Can Auto-Mitigation Help?}
    E -- No --> A
    E -- Yes --> F[Run Safe Mitigation]
    F --> A

Loading

---

🏁 Stay online. Slash downtime. Let AI watch your validator.