Multi-Agent System for Autonomous Ceph Cluster Management

This project implements an intelligent, multi-agent orchestration system for monitoring, analyzing, and managing Ceph storage clusters using CrewAI. At its core, a centralized CephOrchestrator agent coordinates a team of specialized agents to perform real-time cluster health checks, bug triage, documentation lookups, performance monitoring, and automated recommendations.

Key Features:

• 🔍 Cluster Status Evaluation via CephViz Agent
• 📊 Performance & Disk Analysis via Observability Agent
• 🐞 Bug Monitoring via BugIntelligence Agent (Bugzilla Integration)
• 📚 Ceph Docs Lookup via Maverick Agent
• 🧠 Automated Health Recommendations via CephPerf Agent
• 🤖 Hierarchical Task Planning using CrewAI-style orchestration

🧱 Built With:

• Python, CrewAI, LangChain Tools
• Ceph CLI + SSH, Metrics via PostgreSQL, Bugzilla API, Ceph Docs Search

Python - 3.11

1. Install uv package manager: https://docs.astral.sh/uv/getting-started/installation/

2. Sync dependencies:

uv sync

3. Optional - if you want to use python 3.11.x when you have multiple python versions installed.

   uv venv -p 3.11
   source .venv/bin/activate

cd src/
uv run scripts/scrape_ceph_documentation.py

cd src/
uv run agents/maverick/backend/parse_documentation.py

cd src
uv run orchestration/flow.py

cd src
streamlit run frontend/app.py