NVDebug User Guide#

Welcome to the NVDebug user guide! This tool helps you collect diagnostic information from NVIDIA server platforms to troubleshoot issues effectively.

🎯 Quick Navigation

📚 Learning Path

Note

Start with Quick Start → Installation → Basic Usage → Configuration → Advanced Usage

What is NVDebug?

NVDebug is NVIDIA’s diagnostic collection tool that gathers comprehensive system information from:

  • Out-of-band (OOB) data from Baseboard Management Controllers (BMCs)

  • Host system logs and diagnostic information

  • Hardware telemetry and error records

  • Network and storage configuration details

This information helps NVIDIA support teams and system administrators quickly identify and resolve issues.

Note

Supported Platforms: NVDebug works with NVIDIA DGXâ„¢, HGXâ„¢, MGXâ„¢, and GB series platforms. See Supported Baseboards for the complete list.

Key Benefits

✅ Comprehensive Data Collection: Gathers logs from multiple sources in a single command

✅ Standardized Format: Outputs structured data for easy analysis

✅ Remote & Local Operation: Works from remote machines or directly on the target system

✅ Automated Collection: Reduces manual troubleshooting time

✅ Secure: Supports encrypted connections and credential management

Quick Start

Ready to try NVDebug? Here’s the fastest way to get started:

# Basic out of band collection (you'll need your system details)
nvdebug -i <BMC_IP> -u <BMC_USER> -p <BMC_PASS>

# Basic in band collection (you'll need your system details)
nvdebug -I <HOST_IP> -U <HOST_USER> -H <HOST_PASS>

# Basic out of band and in band collection (you'll need your system details)
nvdebug -i <BMC_IP> -u <BMC_USER> -p <BMC_PASS> -I <HOST_IP> -U <HOST_USER> -H <HOST_PASS>

For a complete walkthrough with examples, see Quick Start Guide.

Need Help?

📖 What’s New

  • Updated for latest NVDebug version

  • Simplified configuration examples

  • Configuration with consolidated rack management

  • New quick-start guide

🔗 Related Resources