Skip to navigation

Introduction

View as Markdown
Download PDF

This Upgrade Manual describes the procedure for upgrading a Bright cluster that is being managed by Bright Cluster Manager version 9.2, to a Base Command Manager cluster managed by Base Command Manager version 10.0.

NOTE: Base Command Manager (BCM) is the name given, with release 10.0, to the former Bright Cluster Manager. The word “Bright” may still show up, for example in URLs, paths, and the manuals.

A clean installation of BCM is recommended instead of an upgrade because it is simpler. If an upgrade is to be carried out, then reasonable familiarity with Bright Cluster Manager is assumed, and the Upgrade Manual should be read carefully before the upgrade. A good idea is to look up any unfamiliar concepts described in this manual in the other Bright/BCM manuals.

The BCM Support team can be contacted using https://enterprise-support.nvidia.com/s/create-case if there are BCM-related upgrade issues for which the documentation is insufficient.

1.1 Supported Bright Versions And Linux Distributions

1.1.1 Upgrades From The Following Bright Versions Are Supported:

  • Bright 9.2

1.1.2 The Following Linux Distributions Are Supported:

  • RedHat Enterprise Linux 8 (update 8.8 and higher)

  • RedHat Enterprise Linux 9 (update 9.2 and higher)

  • Rocky Linux 8 (update 8.8 and higher)

  • Rocky Linux 9 (update 9.2 and higher)

  • SuSE Linux Enterprise Server 15 (SLES15 SP4) and higher

  • Ubuntu 20.04

  • Ubuntu 22.04

1.1.3 Upgrading The Underlying OS Is Not Supported

Upgrading the underlying OS, that is, upgrading from an earlier major release of the OS upon which BCM runs, to the next major release of the OS (for example, RHEL8 to RHEL9) while keeping BCM as it is, is not supported. If the OS is to be upgraded, then BCM must be reinstalled from scratch.

1.1.3.1 Exception: DGX OS Upgrade Is Supported, But Not Covered In This Upgrade Manual

A case that can be seen as an exception for upgrading the underlying OS, is upgrading DGX OS 5 to DGX OS 6, or DGX OS 6 to DGX OS 7. DGX OS is a custom OS for the DGX platform. It is based on BCM running on top of Ubuntu. DGX OS package and release upgrades are covered comprehensively in the dedicated DGX OS documentation at:

The latest supported DGX OS version at the time of writing (September 2025) is DGX OS 7.0.

1.2 Parallel Vs In-place Upgrades

There are two kinds of upgrades possible.

  • A parallel upgrade: This is when a clone is made of the running head node. That clone is kept away from the cluster, and it is the clone that is upgraded. The original head node keeps managing the cluster, so that there is not such a severe pressure on getting the upgrade up before a deadline. This means that if there are unexpected issues that take time to resolve during upgrade, then the original cluster remains unaffected. The head node of the original cluster is replaced with the upgraded head node only after the administrator believes that the upgraded head node is ready.

  • An in-place upgrade: This is when an upgrade is carried out using the existing head node or head nodes in a production cluster. The regular nodes need to be powered off when the upgrade is carried out.

Chapter 2 covers the steps and considerations for the upgrade procedure that are specific to the parallel upgrade.

Chapter 3 covers the steps and considerations for the upgrade procedure that are specific to the in-place upgrade.

Download this manual as a PDF