Depl0y

Proxmox VE Management Panel — Infrastructure at your fingertips

⭐ If Depl0y saves you time, give it a star — it helps others find the project!

Depl0y is a free, open-source web control panel for Proxmox VE. Manage VMs, clusters, nodes, storage, backups, networking, firewall, HA, and physical hardware from a single dark-mode interface.

---

Screenshots

Dashboard

Customizable widget grid with drag-and-drop reordering — tiles stack asymmetrically (masonry layout)

Proxmox Datacenters & Hosts

Multi-datacenter view with node health, cluster status, join/unjoin cluster actions, and resource bars

VM Management

Full VM and LXC container management — lifecycle, config, snapshots, migrate, clone, console

Cluster Overview

Cluster-wide status, quorum, node health, HA resources, and replication jobs

Node Monitor

Per-node RRD charts (CPU, RAM, network, disk I/O), task log, storage, and network configuration

Backup Manager

PBS datastore browsing, backup schedule CRUD, and manual backup triggers

Storage

Storage pools across all nodes with content browsing and ISO/cloud image management

Firewall

Cluster, node, and VM-level firewall rules with security groups and IPsets

Replication

Proxmox replication job management with force-sync and log viewer

Federation Map

Live OpenStreetMap showing all datacenters with online/offline pins

iDRAC / iLO Dashboard

Hardware health, power, temperature, and fan monitoring for Dell iDRAC and HPE iLO via Redfish

Deploy LLM

One-click LLM deployment — Ollama, llama.cpp, vLLM, LocalAI with optional GPU passthrough, Open WebUI, and ComfyUI

Users

Role-based access control — Admin, Operator, Viewer — with 2FA/TOTP support

Settings

SMTP, webhooks, cloud image setup, cluster SSH, HA enablement, and one-click system updates

---

Features

🖥️ Full Proxmox VE Management

• VMs — start/stop/reboot/suspend/resume, config editing (CPU, RAM, disks, NICs), snapshots, clone, migrate, firewall, VNC console, QEMU serial terminal
• LXC Containers — lifecycle, config editing, snapshots, terminal (xterm.js)
• Nodes — RRD metrics charts, VM + LXC list, storage browser, network config, task log, node terminal
• Cluster — status, node list, HA groups and resources, quorum monitoring
• Cluster Join / Unjoin — join any node to a cluster (fingerprint auto-fetched), remove nodes from cluster
• Replication — job CRUD, force-sync, log viewer
• Node Evacuation — migrate all VMs off a node to other online nodes
• Firewall — cluster, node, and VM-level rules; security groups; IPsets
• Backup — schedule CRUD, manual trigger, PBS datastore browsing
• Storage — pool management, content browsing, ISO and cloud image management
• Networking — bridge/bond/VLAN config with apply-pending support

📊 Dashboard

• Widget grid — drag-and-drop reordering with masonry (asymmetric column) layout
• 10+ widgets — CPU, RAM, storage, network traffic, disk I/O, VM status, alerts, activity feed, quick actions
• Clickable tiles — every widget links to its management view
• Per-widget refresh — each widget auto-refreshes independently

🗺️ Federation & Multi-Datacenter

• Multi-host — add and manage multiple Proxmox VE hosts with API token or password auth
• Live Map — OpenStreetMap/Leaflet with datacenter pins (blue = online, red = offline)
• Federated Summary — aggregate VM/node/storage stats across all registered hosts
• Federated Dashboard — cross-datacenter VM/node overview in one view

🖧 iDRAC / iLO Out-of-Band Management

• Redfish Dashboard — unified health, power, temperature, wattage for all BMC-equipped servers
• Power Control — on, off, graceful shutdown, force/graceful restart, PXE boot
• Hardware Inventory — CPUs, DIMMs, storage controllers & drives, firmware, NICs, SEL
• Multi-vendor — Dell iDRAC (including iDRAC 8 / 13G) and HPE iLO via Redfish v1

🤖 LLM Deployment

• Simple + Advanced modes — 4 questions to deploy, or full control over engine/model/GPU/OS/storage
• 4 Engines — Ollama, llama.cpp (GGUF), vLLM (OpenAI-compatible), LocalAI (Docker)
• 15+ Models — Llama 3.x, Mistral, Phi-4, Gemma, Qwen, DeepSeek, Code Llama, and more
• GPU Passthrough — NVIDIA (CUDA) and AMD (ROCm) with automatic driver install
• Add-ons — Open WebUI, ComfyUI (Stable Diffusion), AI auto-tuning, RAG, conversation logging

📥 VM Import

• File upload — OVA, OVF, VMDK, VHD, VHDX, QCOW2, RAW via drag & drop
• VMware direct — connect to ESXi or vCenter, browse and pull VMs over the network
• Auto-parse — OVF descriptors extracted for name, CPU, RAM, disk, OS type
• Disk conversion — VMDK/VHD/VHDX → qcow2 via qemu-img automatically

⚡ Cloud Image Deployment

• 30-second deployments — Ubuntu, Debian, Rocky, AlmaLinux (after one-time template setup)
• Cloud-Init — hostname, user, SSH key, static IP, DNS, package injection

🔐 Security & Access Control

• Role-based — Admin, Operator, Viewer with route-level enforcement
• 2FA / TOTP — authenticator app support with QR code setup
• Encrypted storage — all passwords and API tokens encrypted at rest (Fernet)
• Audit log — every user action and system change recorded
• Rate limiting — 100 req/min globally with security headers

🔄 VM Update Management

• One-click updates — check and install updates on any managed Linux VM via SSH
• Real-time streaming — live terminal output as apt/dnf runs
• Auto-scheduled checks — configurable interval (6h–7d) for automatic checks

---

Quick Start

One-Line Installation

curl -fsSL http://deploy.agit8or.net/downloads/install.sh | sudo bash

Installs all dependencies, configures nginx, and creates a systemd service — ready in ~30 seconds.

Post-Installation

1. Open http://your-server-ip — default credentials: admin / admin (change immediately)
2. Enable 2FA — Settings → User Profile → Enable TOTP
3. Add a Proxmox host — Proxmox Hosts → Add Datacenter → test connection
4. Deploy or import a VM

Updating

cd /opt/depl0y && git pull origin main && sudo bash deploy.sh

Or: Settings → System Updates → Check for Updates → Install

---

Configuration

Environment Variables (/etc/depl0y/config.env)

SECRET_KEY=your_jwt_secret_key_minimum_32_chars
ENCRYPTION_KEY=your_fernet_encryption_key
DATABASE_URL=sqlite:////var/lib/depl0y/db/depl0y.db
DEBUG=false
LOG_LEVEL=INFO

Storage Locations

Path	Contents
/var/lib/depl0y/db/depl0y.db	SQLite database
/var/lib/depl0y/isos	ISO images
/var/lib/depl0y/cloud-images	Cloud image templates
/var/lib/depl0y/ssh_keys	SSH key pairs
/var/log/depl0y/	Application logs
/tmp/depl0y-imports/	Temporary VM import working directory

---

Architecture

┌─────────────────────────────────────┐
│         Frontend (Vue.js 3)         │
│   SPA · Axios · Chart.js · Leaflet  │
│   Dark Mode · Widget Grid · xterm   │
└──────────────┬──────────────────────┘
               │ HTTP REST (/api/v1)
┌──────────────▼──────────────────────┐
│       Backend (FastAPI + Python)     │
│  Auth · VMs · Cluster · Import      │
│  LLM · HA · Backup · iDRAC · Alerts │
└──────┬──────────────┬───────────────┘
       │              │
┌──────▼──────┐ ┌────▼────────────────┐
│   SQLite DB  │ │   Proxmox VE API    │
│  (users/VMs/ │ │  nodes/qemu/cluster │
│  settings)   │ │  + Redfish / SSH    │
└─────────────┘ └─────────────────────┘

Key dependencies: proxmoxer, pyVmomi, paramiko, SQLAlchemy, Pydantic, APScheduler, python-jose, Leaflet

---

API Documentation

• Swagger UI: http://your-server/api/v1/docs
• ReDoc: http://your-server/api/v1/redoc
• In-App API Explorer: Sidebar → API Explorer

---

Development

# Backend
cd backend && python3 -m venv venv && source venv/bin/activate
pip install -r requirements.txt
uvicorn app.main:app --reload

# Frontend
cd frontend && npm install && npm run dev

# Production build + deploy
sudo bash /opt/depl0y/deploy.sh

---

Troubleshooting

Cannot connect to Proxmox host

• Verify credentials and network connectivity
• Disable verify_ssl for self-signed certificates
• Ensure Proxmox API port (8006) is reachable

VM import fails

• Ensure local storage exists with enough space on the target node
• SSH must be configured between Depl0y server and Proxmox host (Settings → SSH Setup)

Cluster join fails

• Ensure the root@pam password is correct for the cluster master node
• The fingerprint is auto-fetched — if it fails, fetch it manually: pvecm status on the master

Backend logs

sudo journalctl -u depl0y-backend -f

For more help: GitHub Issues

---

Contributing

Pull requests welcome. For major changes, open an issue first to discuss what you'd like to change.

---

License

MIT — see LICENSE for details.