Nvidia RTX Service Manual
Nvidia RTX Service Manual

Nvidia RTX Service Manual

Blade server

Advertisement

Quick Links

RTX BLADE SERVER
DU-09559-001 _v01 | July 2019
Service Manual

Advertisement

Table of Contents
loading

Summary of Contents for Nvidia RTX

  • Page 1 RTX BLADE SERVER DU-09559-001 _v01 | July 2019 Service Manual...
  • Page 2: Table Of Contents

    TABLE OF CONTENTS Chapter  1.  Introduction..................1 1.1. NVIDIA Enterprise Support Portal..............3 1.2. NVIDIA Enterprise Support Email............... 3 1.3. NVIDIA Enterprise Support - Local Time Zone Phone Numbers........3 Chapter 2. Fan Module Replacement................ 4 2.1. Fan Module Replacement Overview..............4 2.2. Replacing the Faulty Fan Module..............4 Chapter ...
  • Page 3: List Of Figures

    LIST OF FIGURES Figure 1 Front System Components ................2 Figure  2 Rear System Components ................2 www.nvidia.com RTX Blade Server DU-09559-001 _v01 | iii...
  • Page 4 RTX Blade Server DU-09559-001 _v01 | iv...
  • Page 5: Chapter  1.  Introduction

    Be sure to familiarize yourself with the NVIDIA Terms & Conditions documents before attempting to perform any modification or repair to the RTX Blade Server System. These Terms & Conditions for the RTX Blade Server System can be found through the NVIDIA DGX Systems Support page.
  • Page 6 Introduction Figure 1 Front System Components Figure 2 Rear System Components Contact NVIDIA Enterprise Support for replacement instructions and guidance for components that are not included in this document. www.nvidia.com RTX Blade Server DU-09559-001 _v01 | 2...
  • Page 7: Nvidia Enterprise Support Portal

    The best way to file an incident is to log on to the NVIDIA Enterprise Support portal. 1.2. NVIDIA Enterprise Support Email You can also send an email to enterprisesupport@nvidia.com. 1.3. NVIDIA Enterprise Support - Local Time Zone Phone Numbers Visit NVIDIA Enterprise Support (http://www.nvidia.com/en-us/support/enterprise/) www.nvidia.com...
  • Page 8: Chapter 2. Fan Module Replacement

    Chapter 2. FAN MODULE REPLACEMENT The RTX Blade Server fan modules can be replaced while power is applied to the system and the system is running. 2.1. Fan Module Replacement Overview This is a high-level overview of the steps needed to replace the fan modules.
  • Page 9 Quickly insert the new fan module into the empty slot, and continue pushing into the slot until an audible click occurs from the latch on the handle. Caution Replace the fan module within 30 seconds to prevent overheating of the system components. www.nvidia.com RTX Blade Server DU-09559-001 _v01 | 5...
  • Page 10 Fan Module Replacement Confirm that the amber fault light does not light up. www.nvidia.com RTX Blade Server DU-09559-001 _v01 | 6...
  • Page 11: Chapter  3.  Blade Replacement

    Chapter 3. BLADE REPLACEMENT A faulty RTX Blade Server blade can be replaced while power is applied to the system and the system is running. 3.1. Blade Replacement Overview This is a high-level overview of the steps needed to replace the blades.
  • Page 12 Blade Replacement b) Once you locate the RTX Blade Server with the faulty blade, identify the blade to be replaced by the flashing blue and green LEDs. On the faulty blade, push the release latch up to unlock the blade ejector lever.
  • Page 13 Important Do not carry the blade by the lever as this could cause the lever to break. Expect the sudden souind of the door to the slot shutting as you remove the blade. The slot door shuts to divert airflow from the empty slot. www.nvidia.com RTX Blade Server DU-09559-001 _v01 | 9...
  • Page 14 Caution Keep fingers out of the way of the lever as you insert the blade. When the blade meets the midplane connectors in the back, raise the lever back into position to ensure the blade connects with the midplane connectors. www.nvidia.com RTX Blade Server DU-09559-001 _v01 | 10...
  • Page 15 Blade Replacement Make sure you can hear or feel the click as the lever locks into place to finalize the insertion. Press the power button on the blade. www.nvidia.com RTX Blade Server DU-09559-001 _v01 | 11...
  • Page 16: Chapter 4. Chassis Monitoring Module Replacement

    Chapter 4. CHASSIS MONITORING MODULE REPLACEMENT The RTX Blade Server blades must be shut down before replacing the chassis monitoring module. 4.1. Chassis Monitoring Module Replacement Overview This is a high-level overview of the steps needed to replace the chassis monitoring modules (CMM).
  • Page 17 Pull the CMM out of the system. Expect the sudden souind of the door to the CMM bay shutting as you remove the CMM. The bay door shuts to divert airflow from the empty bay. www.nvidia.com RTX Blade Server DU-09559-001 _v01 | 13...
  • Page 18 Confirm all blades are powered off by checking the green LEDs in the front of the chassis are not lit. b) Insert the new CMM into the slot. c) Close the lever to make sure the connectors engage with the midplane and the locking tab clicks into place. www.nvidia.com RTX Blade Server DU-09559-001 _v01 | 14...
  • Page 19 Chassis Monitoring Module Replacement Confirm that the fault light is not lit and that the green light comes up and presents a heartbeat (flashing). Connect the remote management RJ45 cables. www.nvidia.com RTX Blade Server DU-09559-001 _v01 | 15...
  • Page 20: Chapter 5. Switch Module Replacement

    Chapter 5. SWITCH MODULE REPLACEMENT The RTX Blade Server switch module can be replaced while power is applied to the system, but the blades must not be active. 5.1. Switch Module Replacement Overview This is a high-level overview of the steps needed to replace the switch modules.
  • Page 21 Pull the switch module out of the system. Expect the sudden souind of the door to the switch bay shutting as you remove the switch. The bay door shuts to divert airflow from the empty bay. www.nvidia.com RTX Blade Server DU-09559-001 _v01 | 17...
  • Page 22 Install the new switch module. a) Confirm all blades are powered off by checking the green LEDs in the front of the chassis are not lit. b) Insert the new switch module into the slot. www.nvidia.com RTX Blade Server DU-09559-001 _v01 | 18...
  • Page 23 Close the lever to make sure the connectors engage with the midplane and the locking tab clicks into place. Confirm that the fault light is not lit and that the green light comes up and presents a heartbeat (flashing). www.nvidia.com RTX Blade Server DU-09559-001 _v01 | 19...
  • Page 24 Switch Module Replacement Connect the remote management RJ45 cables. www.nvidia.com RTX Blade Server DU-09559-001 _v01 | 20...
  • Page 25: Chapter 6. Power Supply Replacement

    Chapter 6. POWER SUPPLY REPLACEMENT An RTX Blade Server power supply unit (PSU) can be replaced while power is applied to the system and the system is running. 6.1. Power Supply Replacement Overview This is a high-level overview of the steps needed to replace a power supply.
  • Page 26 ‣ If the five remaining PSUs are working and energized, then you do not need to shut down power to the RTX Blade Server System.. ‣ If fewer than five PSUs are working and energized, then you do need to shut down power to the RTX Blade Server System.
  • Page 27 Continue using the black handle to pull the power supply out of the system. Install the new power supply. a) Insert power supply to the chassis and confirm its fully inserted with the audible click that the latch makes. www.nvidia.com RTX Blade Server DU-09559-001 _v01 | 23...
  • Page 28 Power Supply Replacement b) Install the power cord and confirm that the LED turns green. c) If needed, power on the system. www.nvidia.com RTX Blade Server DU-09559-001 _v01 | 24...
  • Page 29 LIFE CRITICAL APPLICATION). NVIDIA EXPRESSLY DISCLAIMS ANY EXPRESS OR IMPLIED WARRANTY OF FITNESS FOR SUCH HIGH RISK USES. NVIDIA SHALL NOT BE LIABLE TO CUSTOMER OR ANY THIRD PARTY, IN WHOLE OR IN PART, FOR ANY CLAIMS OR DAMAGES ARISING FROM SUCH HIGH RISK USES.

Table of Contents