For AI agents: a documentation index is available at the root level at /llms.txt and /llms-full.txt. Append /llms.txt to any URL for a page-level index, or .md for the markdown version of any page.
    • NVIDIA Switch Infrastructure
    • I want to...
  • Quick Start
    • Start Here
    • Getting Started with Config Manager
    • TUI Wizard Reference
    • Configuration Samples
    • Interfaces
    • Local Development Quick Start
    • First Run Tour
  • Config Manager Overview
    • Config Manager Concepts
    • Getting Started with Nautobot
  • User Guides
    • New Site Bringup
    • Workflow Lifecycle
  • Deployment
    • Hosting Options
    • Network Topology Requirements
    • Firewall Ports
    • Airgapped Deployment
    • Troubleshooting
  • Services
      • Network ZTP
      • ZTP Architecture
      • Configuration Guide
      • Upload Images
      • ZTP API
NVIDIANVIDIA
Developer-friendly docs for your API
Privacy Policy | Your Privacy Choices | Terms of Service | Accessibility | Corporate Policies | Product Security | Contact

Copyright © 2026, NVIDIA Corporation.

LogoLogo
On this page
  • Overview
  • System Architecture
  • API Endpoints
  • Authorization Flow
  • Device Endpoint Authorization
  • Admin Endpoint Authorization
  • Provisioning Workflow
  • Storage
  • Security
  • Authentication & Authorization
  • Data Protection
  • Monitoring & Metrics
  • Error Handling
ServicesNetwork ZTP Service

NVIDIA Config Manager ZTP Architecture

||View as Markdown|
Previous

Config Manager Network ZTP

Next

Configuration Guide

Overview

The NVIDIA Config Manager ZTP Server facilitates Zero Touch Provisioning (ZTP) for network devices. It serves boot scripts, rendered configurations, and firmware images during the device provisioning process.

System Architecture

API Endpoints

The API is organized into three main endpoint groups:

  • Device Endpoints: Device-specific operations for boot scripts, configurations, firmware, and provisioning status
  • Firmware Endpoints: Platform/version-based firmware access
  • File Endpoints: Generic file storage and retrieval operations

Authorization Flow

Device Endpoint Authorization

Device-specific endpoints authorize requests in one of two ways:

  1. Device-originated requests must come from an IP address associated with the device in the device management system
  2. User-originated requests must come through the Envoy gateway as an authenticated user when SSO is enabled for the deployment
  3. Unauthorized requests receive a 403 Forbidden response

Admin Endpoint Authorization

Admin endpoints (file uploads, sync triggers) require an authenticated user request through the Envoy gateway when SSO is enabled for the deployment.

Provisioning Workflow

The typical ZTP workflow follows these steps:

Device Boots
│
├─→ DHCP Provides Boot File URL
│
├─→ Device Requests Boot Script
│ └─→ GET /v1/device/{uuid}/boot-script
│
├─→ Device Downloads Firmware
│ └─→ GET /v1/device/{uuid}/firmware
│
├─→ Device Loads Configuration
│ └─→ GET /v1/device/{uuid}/config/{configlet}
│
├─→ Device Validates Serial
│ └─→ POST /v1/device/{uuid}/validate_serial
│
├─→ Device Marks Provisioned
│ └─→ POST /v1/device/{uuid}/provisioned
│ │
│ ├─→ Update Device Status
│ │
│ └─→ Trigger Backup Workflow
│
└─→ Provisioning Complete

Storage

The system uses object storage for firmware and file storage:

  • Files are organized by platform and version
  • Firmware images are tagged for identification
  • Files include SHA256 checksums for verification
  • Large files are streamed efficiently

Security

Authentication & Authorization

  • Device endpoint authorization allows registered devices by source IP and authenticated users through the Envoy gateway when SSO is enabled
  • Admin endpoints require an authenticated user through the Envoy gateway when SSO is enabled
  • Serial number validation prevents unauthorized provisioning

Data Protection

  • Sensitive configuration files are only accessible to registered devices or authenticated users
  • Firmware images are served with checksums for verification

Monitoring & Metrics

The system exposes Prometheus metrics at /metrics:

  • Device request metrics with labels for client IP, base URL, and device UUID
  • Standard HTTP metrics (request duration, count, status codes)

Error Handling

The API uses standard HTTP status codes:

  • 400 Bad Request: Invalid request parameters or body
  • 403 Forbidden: Authorization failure
  • 404 Not Found: Resource not found
  • 500 Internal Server Error: Server error

Error responses include detailed error messages to help diagnose issues.