Skip to content

Pin Dcgm version 4.5.1-1 #5197

Open
saara-tyagi27 wants to merge 8 commits intoGoogleCloudPlatform:developfrom
saara-tyagi27:dcgm-pin-other-version
Open

Pin Dcgm version 4.5.1-1 #5197
saara-tyagi27 wants to merge 8 commits intoGoogleCloudPlatform:developfrom
saara-tyagi27:dcgm-pin-other-version

Conversation

@saara-tyagi27
Copy link
Contributor

@saara-tyagi27 saara-tyagi27 commented Feb 5, 2026

Summary

Pin DCGM version to latest 4.5.1 version

Submission Checklist

NOTE: Community submissions can take up to 2 weeks to be reviewed.

Please take the following actions before submitting this pull request.

  • Fork your PR branch from the Toolkit "develop" branch (not main)
  • Test all changes with pre-commit in a local branch #
  • Confirm that "make tests" passes all tests
  • Add or modify unit tests to cover code changes
  • Ensure that unit test coverage remains above 80%
  • Update all applicable documentation
  • Follow Cluster Toolkit Contribution guidelines #

@gemini-code-assist
Copy link
Contributor

Summary of Changes

Hello @saara-tyagi27, I'm Gemini Code Assist1! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!

This pull request enhances the robustness of the HPC Toolkit by introducing comprehensive validation for Google Cloud reservations, ensuring that resources specified in blueprints are correctly configured and accessible. Additionally, it standardizes the version of the Datacenter GPU Manager in a key machine learning blueprint, contributing to deployment stability.

Highlights

  • DCGM Version Pinning: The datacenter-gpu-manager packages are now explicitly pinned to version 1:4.5.1-1 and held to prevent unintended upgrades in the a3ultra-slurm-blueprint.yaml example.
  • Google Cloud Reservation Validation: New Go functions have been introduced to validate the existence and correct zone of Google Cloud reservations, providing detailed diagnostics and hints if a reservation is not found where expected.
Changelog
  • examples/machine-learning/a3-ultragpu-8g/a3ultra-slurm-blueprint.yaml
    • Updated apt install commands to specify version 1:4.5.1-1 for datacenter-gpu-manager-4-cuda12, datacenter-gpu-manager-4-dev, and datacenter-gpu-manager-4-core.
    • Added apt-mark hold commands for the aforementioned DCGM packages to prevent automatic updates.
  • pkg/validators/cloud.go
    • Imported the regexp package.
    • Defined reservationNameRegex and resKeyRegex for pattern matching reservation identifiers.
    • Implemented findReservationInOtherZones to list zones where a given reservation name exists.
    • Implemented TestReservationExists to perform a detailed check for a reservation's presence in a specific project and zone, including handling access issues, cross-zone searches, and providing HintError for misconfigured zones.
    • Implemented testReservationExists to parse reservation input, determine the owning project and target name, and call TestReservationExists.
  • pkg/validators/validators.go
    • Added testReservationExistsName constant.
    • Registered testReservationExists in the implementations map.
    • Modified the defaults function to automatically add testReservationExists validators for blueprint variables whose names match resKeyRegex.
  • pkg/validators/validators_test.go
    • Added new config.Validator definitions (resInp, resExists, myResInp, myResExists) for testing reservation validation.
    • Extended TestDefaultValidators with test cases to verify that testReservationExists is correctly included in the default validators when reservation_name or my_reservation variables are present in the blueprint.
  • tools/validate_configs/validate_configs.sh
    • Appended test_reservation_exists to the VALIDATORS_TO_SKIP environment variable to exclude it from certain validation runs.
Activity
  • The pull request includes a standard submission checklist, indicating that the author is following community contribution guidelines. This involves steps like forking from the develop branch, running pre-commit hooks, ensuring all tests pass, adding or modifying unit tests for coverage, and updating documentation.
Using Gemini Code Assist

The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.

Invoking Gemini

You can request assistance from Gemini at any point by creating a comment using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands on the current page.

Feature Command Description
Code Review /gemini review Performs a code review for the current pull request in its current state.
Pull Request Summary /gemini summary Provides a summary of the current pull request in its current state.
Comment @gemini-code-assist Responds in comments when explicitly tagged, both in pull request comments and review comments.
Help /gemini help Displays a list of available commands.

Customization

To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.

Limitations & Feedback

Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here.

You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension.

Footnotes

  1. Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution.

Copy link
Contributor

@gemini-code-assist gemini-code-assist bot left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Code Review

This pull request contains two significant changes. Firstly, it pins the datacenter-gpu-manager package to version 4.5.1-1 in the A3 UltraGPU Slurm example, which improves environment reproducibility. Secondly, it introduces a new validator for checking the existence and location of Google Cloud reservations. This validator is well-implemented with robust error handling and is automatically applied to relevant blueprint variables.

The code quality is excellent and includes appropriate tests for the new functionality.

One point of feedback on the process: the pull request title and description only mention the first change. To better reflect the scope of work, I recommend updating them to include the addition of the reservation validator.

@saara-tyagi27 saara-tyagi27 changed the title Dcgm pin 4.5.1-1 version Pin Dcgm version 4.5.1-1 Feb 5, 2026
@saara-tyagi27 saara-tyagi27 force-pushed the dcgm-pin-other-version branch 2 times, most recently from 1dd2afb to 216090a Compare February 5, 2026 11:03
@saara-tyagi27 saara-tyagi27 marked this pull request as ready for review February 17, 2026 06:14
@saara-tyagi27 saara-tyagi27 requested review from a team and samskillman as code owners February 17, 2026 06:14
@saara-tyagi27 saara-tyagi27 added the release-improvements Added to release notes under the "Improvements" heading. label Feb 17, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

release-improvements Added to release notes under the "Improvements" heading.

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants